UNIVERZITET U BEOGRADU 
 
SAOBRAĆAJNI FAKULTET 
 
 
 
 
 
 
 
 
Bojana D. Mirković 
 
 
BALANSIRANO KORIŠĆENJE 
KAPACITETA I PLANIRANJE RAZVOJA 
ELEMENATA AERODROMA 
 
 
doktorska disertacija 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Beograd, 2013 
 
 
UNIVERSITY OF BELGRADE 
 
FACULTY OF TRANSPORT AND TRAFFIC 
ENGINEERING 
 
 
 
 
 
 
 
 
Bojana D. Mirković 
 
 
AIRPORT AIRSIDE BALANCED 
CAPACITY USAGE AND PLANNING 
 
 
Doctoral Dissertation 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Belgrade, 2013
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
i 
 
 
 
 
Mentor:  
Prof. dr Vojin Tošić, redovni professor (u penziji), Univezitet u Beogradu – Saobraćajni 
fakultet 
 
Članovi komisije: 
Prof. dr Obrad Babić, redovni profesor, Univezitet u Beogradu – Saobraćajni fakultet 
(predsednik komisije) 
Prof. dr S. Chan Wirasinghe, redovni profesor, Tehnički fakultet, Univerzitet u 
Kalgariju, Kanada 
Dr Milan Janić, viši istraživač, Institut za Transport, Tehnički univerzitet u Delftu, 
Holandija; naučni savetnik, Univerzitet u Beogradu – Saobraćajni fakultet 
Prof. dr Milica Kalić, redovni profesor, Univezitet u Beogradu – Saobraćajni fakultet 
 
 
 
 
 
 
 
Datum odbrane: 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
ii 
 
 
 
 
Adviser:  
Dr Vojin Tošić, Professor (retired) at University of Belgrade – Faculty of Transport and 
Traffic Engineering, Serbia 
 
Committee: 
Dr Obrad Babić, Professor at University of Belgrade – Faculty of Transport and Traffic 
Engineering, Serbia (Chairman) 
Dr S. Chan Wirasinghe, Professor at Schulich - School of Engineering, University of 
Calgary, Alberta, Canada  
Dr Milan Janic, Senior Researcher at Transport Institute, Delft University of 
Technology, Netherlands; and Research Professor at University of Belgrade – Faculty 
of Transport and Traffic Engineering, Serbia 
Dr Milica Kalić, Professor at University of Belgrade – Faculty of Transport and Traffic 
Engineering, Serbia 
 
 
 
 
 
 
Public defense: 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
iii 
 
 
 
 
Zahvalnica: 
 
Zahvaljujem se svojoj porodici, mentoru, prijateljima i kolegama, i svima ostalima koji 
su mi pružili veliku podršku i verovali u mene, čak i kada sama nisam. Želim posebno 
da izdvojim one koji su mi, svesno ili nesvesno, ulili dodatnu snagu kada mi je to bilo 
najviše potrebno: mama i tata, Bata, Ujka, Dragana, Mihael i Jovana. Zahvalnost 
dugujem i odličnom radnom okruženju u mom rodnom gradu Ljigu. I na kraju, ali ne po 
prioritetu, zahvaljujem se svim članovima komisije na savetima, predlozima, pa i 
primedbama, koje su, nadam se, pozitivno uticale na kvalitet konačne verzije moje 
disertacije.  
Izrada ove disertacije je podržana od strane Ministarstva prosvete, nauke i tehnološkog 
razvoja, Republike Srbije, kroz dva naučno-istraživačka projekta Instituta Saobraćajnog 
fakulteta:  
- Predlog metoda za vrednovanje razvojnih scenarija sistema vazdušnog 
saobraćaja u Srbiji (prevozioci, aerodromi i kontrola letenja) sa aspekta 
bezbednosti, efikasnosti, ekonomičnosti i uticaja na životnu sredinu (broj 
15023), 2008-2011; i  
- Podrška održivom razvoju sistema vazdušnog saobraćaja Republike Srbije (broj 
TR36033), 2011-2014. 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
iv 
 
 
 
 
Acknowledgement: 
 
I would like to thank to my family, adviser, friends and colleagues who supported me 
and believed in me. I owe special gratitude to several people who gave me additional 
strength during the toughest last couple of months: my mom and dad, Bata, Ujka, 
Dragana, Michael and Jovana. I also have to mention great working environment in my 
hometown of Ljig. Last, but not least, I am grateful for comments and suggestion 
provided by all the Committee members. I hope I have managed to improve the quality 
of the thesis by incorporating them into the final version. 
The thesis was supported by the Ministry of Education, Science and Technological 
Development, Republic of Serbia, through two projects implemented by the Faculty of 
Transport and Traffic Engineering:  
- Project TR15023, 2008-2010, Methods for Evaluation of Development 
Scenarios of the Air Transport System in Serbia (Airlines, Airports, ATC) – 
Safety, Efficiency, Economic and Environmental Aspects; and  
- Project TR36033, 2011-2014, A Support to Sustainable Development of the 
Republic of Serbia’s Air Transport System.  
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
v 
 
Balansirano korišćenje kapaciteta i planiranje razvoja elemenata aerodroma 
Rezime: 
U doktorskoj disertaciji je predložen postupak za analizu kapaciteta vazdušne strane 
aerodroma, za zadata fizička i operativna ograničenja, i zadate karakteristike potražnje. 
Ovaj postupak podrazumeva povezivanje (postojećeg) modela za procenu kapaciteta 
sistema poletno-sletnih staza sa (proširenim) modelom za procenu kapaciteta  
pristanišne platforme, kroz njihovu funkcionalnu vezu. 
Cilj ove doktorske disertacije je bio vrednovanje i, po potrebi, modifikovanje i 
proširenje postojećih modela za procenu kapaciteta platforme, kao i  definisanje 
funkcionalne veze između poletno-sletne staze i platforme za različite tipove saobraćaja.  
Postojeći modeli su prošireni tako da uzimaju u obzir ograničenja po tipu aviona i 
korisnicima (npr. aviokompanije), kao i po vrsti saobraćaja. U cilju analize osetljivosti, 
predlažene su obvojnice za prikazivanje kapaciteta platforme određene konfiguracije, u 
zavisnosti od strukture potražnje u odnosu na glavne uticajne faktore. 
Analiza je obuhvatila dva osnovna tipa aerodroma sa aspekta njihove uloge u mrežama 
vazdušnog saobraćaja, a to su:  izvorno-ciljni aerodromi, sa dominantnim saobraćajem 
od-tačke-do-tačke, i hub aerodromi, sa dominantnim transfernim saobraćajem za koji je 
karakteristično da se koncentriše u talase. Dodatno su analizirani i aerodromi na kojima 
postoje oba tipa saobraćaja.   
Rezultati disertacije pokazuju da se za izvorno-ciljne aerodrome može koristiti 
standardni pristup prilikom analize ukupnog kapaciteta vazdušne strane aerodroma, u 
kome se poletno-sletna staza i pristanišna platforma posmatraju odvojeno, pri čemu 
manji kapacitet nameće ograničenje ukupnog kapaciteta. Sa druge strane, u slučaju hub 
aerodroma kapacitet platforme i kapacitet poletno-sletne staze se ne mogu posmatrati 
nezavisno jedan od drugog.  
S tim u skladu, u ovoj doktorskoj disertaciji predložen je model za procenu kapaciteta 
platforme na hub aerodromima, koji pored konfiguracije platforme i strukture potražnje 
uzima u obzir i  kapacitet poletno-sletne staze, kao i parametre koji opisuju talasnu 
strukturu saobraćaja. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
vi 
 
Kroz pažljivo definisane primere analizirana je razlika u kapacitetu izvorno-ciljnih i hub 
aerodroma pod istim uslovima, kao i razlika između teoretskog kapaciteta koji hub 
aerodrom može da ponudi koordinisanim letovima i iskorišćenog kapaciteta za zadatu 
talasnu strukturu potražnje. Takođe, kroz primere je prikazana i promena iskorišćenja 
raspoloživog kapaciteta na hub aerodromu ukoliko bi se dopustilo korišćenje slobodnih 
resursa i drugim, osim koordinisanim letovima. Dva scenarija su razmatrana: 
„preferentni“, u kome se dozvoljava korišćenje svih parking pozicija ostalim letovima 
kada ih ne koriste koordinisani letovi, i „eksluzivni“, koji podrazumeva korišćenje 
kontaktnih parking pozicija samo za koordinisane letove. Na posebnim primerima se 
analizira i uticaj promena kapaciteta poletno-sletne staze na kapacitet platforme na hub 
aerodromu, pod različitim uslovima, kako bi se istakle okolnosti pod kojima je ovaj 
uticaj jasno vidljiv, a pod kojima ostaje sakriven.  
Na kraju, u disertaciji se razmatra i pitanje rezervnog/latentnog kapaciteta platforme, 
kao potencijala da se prihvati dodatna potražnja, pre odluke o fizičkom proširenju za 
obezbeđenje dodatnog kapaciteta. Rezervni kapacitet se definiše u funkciji fleksibilnosti 
kapaciteta platforme. U  disertaciji je predloženo izražavanje fleksibilnosti kapaciteta 
platforme, sa zadatim operativnim ograničenjima, u odnosu na apsolutno fleksibilnu 
platformu, koja nema operativnih ograničenja u korišćenju parking pozicija. 
 
 
 
 
Ključne reči: vazdušna strana aerodroma, aerodromska platforma, analitički modeli, 
kapacitet u uslovima zasićenja, funkcionalna veza poletno-sletna staza - platforma, 
izvorno-ciljni aerodrom, hub aerodrom, analiza osetljivosti 
Naučna oblast: Saobraćajno inženjerstvo 
Uža naučna oblast: Aerodromi i bezbednost vazdužne plovidbe 
UDK broj: 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
vii 
 
Airport airside balanced capacity usage and planning 
Abstract: 
The thesis proposes an approach to analyzing the capacity of the existing (built) system 
under given physical and operational constraints and for given demand characteristics. 
The approach considers the linking of the (existing) runway capacity model with the 
(extended) apron capacity model, through the runway-apron functional relationship. 
The objective of the thesis was to evaluate and, if necessary, to modify/expand the 
existing apron capacity estimation models, as well as to define functional relationship 
between the runway system and apron(s).   
Existing apron capacity models are modified to include constraints on both aircraft 
classes and users (e.g. airlines), considering also different traffic types. The thesis also 
suggests apron capacity envelopes to illustrate sensitivity of apron capacity to changes 
in the demand structure with respect to dominant users, provided for a given apron 
configuration.  
Two general airport categories with respect to the role of the airport in the air transport 
network are analyzed: origin-destination airports (serving primarily point-to-point 
flights) and hub airports (serving primarily airline/alliance coordinated flights). 
Furthermore, the thesis also considers the co-existence of point-to-point and coordinated 
flights at a single airport. 
The results of the thesis show that the common approach in the overall airside capacity 
analysis can be applied at origin-destination airports: the runway system and apron(s) 
can be observed independently of each other, deriving the conclusion on the overall 
airside capacity by comparing the two. On the other hand, the finding of the thesis is 
that capacities of the runway system and apron(s) at the hub airports have to be 
observed linked to each other.  
Consequently, a model to estimate apron capacity at hub airport is offered in the thesis. 
In addition to apron configuration and demand structure it also takes into consideration: 
hubbing parameters and the runway system performance.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
viii 
 
Different examples are used to analyze the following: the difference between apron 
capacity for origin-destination airport and hub airport, under the same conditions; the 
difference between the theoretical capacity hub airport can offer to coordinated flight 
and the utilized capacity under given wave-system structure; and utilization of the 
capacity at hub airports when point-to-point traffic is allowed to use idle stands (two 
scenarios are compared: preferential, assuming that contact stands are available to point-
to-point flights between waves, and exclusive, assuming that contact stands are used 
only by coordinated flights). Additional examples are used to analyze the influence of 
the runway system performance on apron capacity at hub airports, under different 
conditions. The aim was to differentiate between the cases where this influence is 
obvious, from the other cases where it is concealed. 
The thesis also addresses the issue of reserve/latent apron capacity, as a potential 
solution for accepting additional demand prior to physical expansion. Reserve capacity 
is defined and discussed through apron capacity flexibility, which is expressed in this 
thesis relative to apron capacity of the absolutely flexible apron (with no operational 
constraints on stand usage). 
 
 
 
 
 
Key words: airport airside, airport apron, analytical modeling, saturation capacity, 
runway-apron functional relationship, origin-destination airport, hub airport, sensitivity 
analysis 
Scientific field: Transport and Traffic Engineering 
Specific scientific field: Airports and Air Traffic Safety 
UDK: 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
ix 
 
Table of contents 
 
Introduction ...................................................................................................................... 1 
1 Airport elements usage with respect to different users ........................................... 4 
1.1 Market segments ............................................................................................ 4 
1.2 Aircraft classes .............................................................................................. 8 
1.3 Airlines/alliances and flight origin/destination .............................................. 8 
1.4 Impact of different user constraints on airport airside capacity .................. 13 
2 Airport categorizations with respect to traffic parameters .................................... 14 
2.1 General ......................................................................................................... 14 
2.2 The role of airports in air transport networks .............................................. 15 
2.2.1 Selected airport categorizations  
 with respect to role of airport in the network .................................. 17 
2.3 Market segments differentiation .................................................................. 18 
2.4 Other considerations .................................................................................... 19 
3 Airport airside capacity - definitions and factors that affect it.............................. 20 
3.1 Definitions of airside capacity ..................................................................... 20 
3.2 Factors that affect airside capacity .............................................................. 22 
3.3 Characteristics of hub airports with  
 implication on airport airside capacity ........................................................ 26 
3.3.1 Spatial and temporal concentration at hub airports ......................... 26 
3.3.2 Ideal wave ........................................................................................ 27 
3.3.3 Hubbing performance ...................................................................... 29 
3.3.4 Wave-system structure .................................................................... 30 
3.3.5 Evolutionary phases ........................................................................ 32 
3.4 Peak period .................................................................................................. 36 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
x 
 
4 Airport airside modeling - literature review ......................................................... 42 
4.1 Airport airside capacity models for supporting planning studies ................ 42 
4.2 Apron capacity modeling ............................................................................. 46 
4.3 A way forward ............................................................................................. 54 
5 An approach to analyzing overall airport airside capacity .................................... 55 
5.1 Conceptual model ........................................................................................ 57 
5.2 Apron capacity estimation module .............................................................. 59 
5.3 Apron capacity representation ..................................................................... 62 
5.4 Utilization factor and stand occupancy times .............................................. 70 
5.5 Runway-apron functional relationship under different traffic patterns ....... 74 
5.5.1 Origin-destination airports .............................................................. 75 
5.5.2 Hub airports – only coordinated flights ........................................... 78 
5.5.3 Hub airports – mix of coordinated and point-to-point flights ......... 91 
5.5.4 The impact of runway system performance on apron capacity ..... 101 
5.6 Reserve apron capacity .............................................................................. 103 
5.7 Possible areas of application ...................................................................... 110 
6 Conclusion .......................................................................................................... 112 
7 List of abbreviations ............................................................................................ 116 
8 List of variables and parameters ......................................................................... 118 
9 References ........................................................................................................... 123 
10 Bibliography ........................................................................................................ 134 
11 Appendices .......................................................................................................... 143 
Appendix 1 – European air transport networks .................................................. 144 
Appendix 2 – Airport categorizations accounting for hubbing ........................... 151 
Appendix 3 – European seasonal airports ........................................................... 156 
Appendix 4 – Minimum connecting times – examples at the world’s airports .. 158 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
xi 
 
Appendix 5 – Indirect connectivity indices ........................................................ 159 
Appendix 6 – Evolution of wave-system in 1990s: 
Example of Munich Airport ......................................................... 161 
Appendix 7 – Selected definitions of peak hour and design day ........................ 163 
Appendix 8 – Custer analysis of the world’s airports 
with respect to traffic parameters ................................................. 165 
Appendix 9 – Models for optimizing available resources utilization ................. 168 
Appendix 10 – Comparison of the analytical model 
and FAA graphical method ........................................................ 170 
Appendix 11 – Apron with flexible layout with respect to aircraft classes ........ 173 
Appendix 12 – Ideal wave with intercontinental and continental flights ........... 174 
Appendix 13 – An impact of the runway system performance 
on apron capacity ....................................................................... 176 
Appendix 14 – Apron capacity estimation tool – basic user guide ..................... 178 
Appendix 15 – Stand allocation at Zurich Airport 
after Switzerland entry into the Schengen zone ........................ 182 
12 Biography ............................................................................................................ 184 
 
 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
xii 
 
List of Figures 
 
Figure 1. Airport users with respect to type of traffic (Mirkovic and Tosic, 2012)...... 5 
Figure 2. Aprons at Munich Airport, Germany (DFS, 2004) ....................................... 6 
Figure 3. Layout of the terminal complex at San Francisco Airport, 
United States (SFO, 2012) ............................................................................. 9 
Figure 4. Layout of the terminal complex at  
 Zurich Airport, Switzerland (ZRH, 2012a) ................................................. 10 
Figure 5. Layout of the terminal complex at  
 Munich Airport, Germany (MUC, 2013) .................................................... 11 
Figure 6. Layout of the terminal complex at  
 John F. Kennedy Airport, United States (JFK, 2012) ................................. 12 
Figure 7. A wave of flights at a hub airport (Dennis, 1994) ....................................... 26 
Figure 8. Ideal wave (Bootsma, 1997, cited in: Burghouwt, 2007) ............................ 27 
Figure 9. Daily traffic at Milano Malpensa, Italy,  
 19th January 2005 (Danesi, 2006) ............................................................... 31 
Figure 10. Daily traffic at London Heathrow, United Kingdom, 
19th January 2005 (Danesi, 2006) ............................................................... 32 
Figure 11. Type 1, O/D airport: Tunis Airport, Tunis (Kraus and Koch, 2006) ........... 33 
Figure 12. Type 2, O/D airport with first hub characteristics: 
Johannesburg  Airport, South Africa (Kraus and Koch, 2006) ................... 33 
Figure 13. Type 3, long-haul hub: Doha Airport, Qatar (Kraus and Koch, 2006) ........ 34 
Figure 14. Type 4, mature hub: Paris Charles de Gaulle Airport, France 
(Kraus and Koch, 2006) ............................................................................... 35 
Figure 15. Type 5, rolling (continuous) hub: Dallas Fort Worth Airport, 
United States (source: Kraus and Koch, 2006) ............................................ 35 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
xiii 
 
 
Figure 16. An approach to identifying airport types  
 based on traffic parameters (Oettl and Reeb, 2012) .................................... 38 
Figure 17. Traffic parameters (Oettl and Reeb, 2012) .................................................. 39 
Figure 18. Fill factor and relative load (Oettl and Reeb, 2012) .................................... 39 
Figure 19. FAA’s graphical method for calculating gate capacity ............................... 50 
Figure 20. An approach to overall airside capacity analysis ......................................... 58 
Figure 21. Example of the runway capacity envelope for a single runway .................. 63 
Figure 22. Apron capacity envelope in Example 1 ....................................................... 64 
Figure 23. Set of apron capacity envelopes illustrating  
 an increase in number of mixed-use stands in Example 1 ........................... 65 
Figure 24. Set of apron capacity envelopes illustrating  
 an increase in number of exclusive-use stands in Example 1 ...................... 65 
Figure 25. Relevant segments of apron capacity envelopes 
for scenarios analyzed in Example 2 ........................................................... 69 
Figure 26. Runway capacity envelope in Example 3 .................................................... 77 
Figure 27. Parameters describing wave-system structure  
 in the case of split waves (WRC≥WL) ........................................................ 79 
Figure 28. Dependence of the wave length on the number of aircraft per wave 
and MCT, assuming runway capacity of 30 arrivals/h ................................ 80 
Figure 29. Dependance of Nlos on MaxCT and MCT,  
 assuming runway capacity of 30 arrivals/h ................................................. 82 
Figure 30. Parameters describing wave-system structure  
 in the case of overlapping waves, WRC<WL ............................................. 83 
Figure 31. Parameters describing wave-system structure  
 in the case with entirely overlapped waves, WRC=SOT ............................ 83 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
xiv 
 
 
Figure 32. Theoretical vs. utilized capacity for different WRCs in Example 3a .......... 87 
Figure 33. Maximum number of aircraft per wave 
for different MaxCTs in Example 3a ........................................................... 87 
Figure 34. Maximum number of aircraft per wave for different MaxCTs 
in Example 3a (contact and remote stands) ................................................. 89 
Figure 35. Apron capacities for O/D and hub airports (Examples 3 and 3a) ................ 90 
Figure 36. Apron capacity sensitivity to share of coordinated and other flights 
under exclusive and preferential use cases in Example 4 ............................ 97 
Figure 37. Relationship between number of coordinated and other aircraft 
handled on the apron, during the WRC, for different demand structures, 
under exclusive and preferential use cases in Example 4 ............................ 98 
Figure 38. Apron capacities for different airport types  
 (Examples 3a and 3b and modified Example 3) .......................................... 99 
Figure 39. Apron capacity sensitivity to runway system performance, 
comparison between Example 3 and Example 5 ....................................... 102 
Figure 40. Apron capacity envelope for Example 1, scatter diagram ......................... 106 
Figure 41. Apron capacity envelope for Example 2a, scatter diagram ....................... 107 
Figure 42. Apron capacity envelopes for a constraint-free apron 
and current configuration in Example 1 .................................................... 108 
Figure 43. Apron capacity envelopes for a constraint-free apron 
and current configuration in Example 2a .................................................. 108 
 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
xv 
 
List of Tables 
 
Table 1. Airports in Europe grouped by dominant market segment 
(source: Eurocontrol, 2007) ........................................................................... 5 
Table 2. Airport categorization based on spatial and temporal concentration 
of traffic (Burghouwt, 2007) ....................................................................... 17 
Table 3. Factors that affect runway system and apron capacity ................................ 25 
Table 4. The first thee representative peaks for each airport category 
(Oettl and Reeb, 2012) ................................................................................. 40 
Table 5. Summary of the relevant literature on apron modeling ............................... 53 
Table 6. Demand structure and average stand occupancy times in Example 2 ......... 68 
Table 7. Current and future demand structure with respect 
to fleet mix and users in Example 2 ............................................................ 68 
Table 8. Apron capacity changes (in %) relative to the current 
apron capacity level (10.6 aircraft/h) in Example 2 .................................... 70 
Table 9. Minimum ATC separation rules between arriving aircraft 
on approach in Example 3 ........................................................................... 76 
Table 10. Minimum ATC separation rules between departing aircraft 
in Example 3 ................................................................................................ 76 
Table 11. Apron configuration, demand structure, turnaround times 
and separation times in Example 3 .............................................................. 77 
Table 12. Apron configuration, demand structure, turnaround times 
and separation times in Example 3b ............................................................ 95 
Table 13. Estimated number of coordinated and other aircraft handled 
on the apron, during the WRC, for different demand structures, 
under exclusive and preferential use cases in Example 4 ............................ 99 
Table 14. Demand structure for pure hub (Example 3a), 
for hub with additional point-to-point traffic 
(Example 3b) and for O/D airport (modified Example 3) ......................... 100 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
1 
 
Introduction 
Airport capacity planning and management is a very challenging task that requires 
matching capacity (which is neither easy, nor cheap to expand) to fast growing demand 
which from time to time suffers changes in its characteristics. We are witnessing the 
transformation from point-to-point networks to hub-and-spoke networks and back, of a 
new user profile appearance (low-cost), the introduction of new large aircraft into 
service (A380), etc.  
The runway system is considered to be the main airport capacity planning parameter. 
This is reasonable, considering that building a new runway is a huge infrastructural 
project for an airport, both in terms of investment, and in terms of capacity gain (in 
aircraft operations per hour). Another huge infrastructural project is the development of 
the terminal complex, when the issue of passenger terminal capacity comes to the fore.  
In order to provide efficient functioning of the whole system between major 
infrastructural changes, it is important that the development of other elements (taxiway 
system, aprons, curb-side, etc.) follows expected demand volume and structure changes. 
Development of other airport elements can be planned and realized in smaller steps, 
which is associated with smaller investments, as well as a lower risk of significant 
capacity/demand mismatch occurrence. However, that is not a justification for these 
issues to be put aside, or to be very poorly considered.  
The objective of the thesis is to evaluate and, where necessary, to modify/expand 
existing capacity estimation models (of the runway system, aprons and entire airside), 
with the aim of suggesting an approach for overall airside capacity analysis.  
In the process of airside capacity analysis it is important to understand the system as a 
whole, accounting for physical and operational constraints on airside elements usage 
and functional relationship between elements. Chapter 1 of the thesis summarizes 
airport elements usage with respect to different criteria: market segments, physical 
constraints and operational constraints, together with their impact on the capacity of the 
entire system. The thesis primarily addresses passenger airports which account for a 
large share of all market segments.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
2 
 
At passenger airports runway-apron functional relationship is mainly affected by the 
role of the airport in the air transport network. Chapter 2 discusses different airport 
categorizations with respect to traffic parameters. It places a special emphasis to airport 
types with respect to their role in the network.  
Chapter 3 summarizes definitions related to airside capacity and the main factors that 
affect it. Special emphasis is given to characteristics of hub airports with implications 
on airside capacity utilization, primarily: temporal concentration, ideal wave, wave-
system structure and evolutionary phases. This chapter also discusses the issue of peak 
period. This includes traditional approaches to derive peak hour and design day, 
together with some of the latest research aimed at deriving typical representative peaks 
for large international airports, based on the selected traffic parameters. 
Chapter 4 provides an overview of the broader literature related to the models for airside 
capacity assessment and analysis, used for supporting planning studies with a medium- 
to long-term horizon. A detailed review of the apron capacity modeling is given in the 
second part of this chapter. It is identified as a research field rather modestly addressed 
in literature, in comparison to very mature fields such as runway system modeling and 
simulation modeling of the entire airside. 
Chapter 5 proposes an approach for overall airside capacity analysis (Section 5.1). The 
approach considers linking (one of the) existing runway system capacity models with 
the apron capacity model (newly developed or modified), through the functional 
relationship between airside elements (the runway system and aprons), taking into 
account terminal airspace and passenger terminal processes, to the extent that is 
necessary.  
In Section 5.2 an extension of existing analytical models is proposed, followed (in 
Section 5.3) by apron capacity representation, using the apron capacity envelope to 
illustrate how the capacity of a certain apron configuration reacts to changes in demand 
structure, with respect to dominant users. The significant variables that influence 
estimated capacity are discussed in Section 5.4. 
Section 5.5 addresses the difference in capacity utilization of available resources, 
depending on the nature of the traffic at the airport. Two main airport types, with 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
3 
 
respect to their role in air transport network, are analyzed: origin-destination airports, 
serving primarily point-to-point traffic (resulting in rather “uniform” traffic distribution 
during the day, with more or less pronounced peak periods) and hub airports serving 
primarily transfer traffic typically concentrated in waves/banks of flights. Following 
network evolution, the thesis also addresses the co-existence of coordinated and point-
to-point flights at a single airport. 
The thesis also addresses the issue of reserve/latent apron capacity (Section 5.6). It is 
discussed through the concept of apron capacity flexibility. On one hand, apron capacity 
flexibility is observed as an indicator of ability of the apron(s) to respond on changes in 
demand structure. At the same time, it is an indicator of latent/reserve capacity that may 
be “activated” by reallocation of available resources i.e. relaxation of the constraints on 
stand use. 
In addition (Section 5.7) a short discussion is given on possible areas of application of 
the proposed approach in the field of airport airside capacity analysis.  
Concluding remarks are summarized in Chapter 6 of the thesis. 
The thesis contains 15 appendices that additionally support certain discussions from the 
main content. Appendix 14 is an introductory guide for an apron capacity estimation 
tool provided as a supplementary material to the thesis.   
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
4 
 
1 Airport elements usage with respect to different users 
The airport system consists of two sub-systems depending on what/who is served by the 
system elements. The airside sub-system primarily serves aircraft (and airport ground 
service vehicles). It considers runway system, taxiway system and apron(s)1, also 
referred (all together) as the airfield. The landside sub-system serves passengers and 
vehicles. It considers terminal buildings, access roads, curbside, parking, etc. The 
airside and landside sub-systems are connected through terminal complex i.e. through 
terminal building-apron link.  
The thesis observes airport airside, while the landside is taken into consideration only to 
the extent to which it is inseparably connected to the airside, e.g. it may impose certain 
constraints on apron usage.  
This chapter summarizes usage of airport airside elements depending on airport users 
(Mirkovic 2011a; Mirkovic and Tosic, 2012; Mirkovic and Tosic, 2013). Users are 
classified with respect to different criteria. Section 1.1 discusses airport elements usage 
with respect to different market segments. Section 1.2 addresses aircraft classes and 
physical constraints. Section 1.3 addresses operational constraints related to 
airline/traffic characteristic. The impact of different user constraints on airside capacity 
is summarized in Section 1.4. 
1.1 Market segments 
When observed in the broadest sense, the users of the airport are associated with various 
purposes of flying: passenger, cargo, military, business aviation, and other (pilot 
education, agricultural flying, panoramic flying etc.), summarized in Figure 1.  
Eurocontrol (2007) differs between seven market segments: traditional scheduled, low-
cost, non-scheduled, business aviation, cargo, military, and other, as given in Table 1. 
The first three fall under passenger market segments.  
                                                 
1 Hereinafter, the term apron(s) is used to represent all aircraft stands, including contact aircraft stands 
(also referred to as gates) and remote stands (also referred to as open stands). 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
5 
 
 
 
Figure 1. Airport users with respect to type of traffic (Mirkovic and Tosic, 2012) 
Table 1. Airports in Europe grouped by dominant market segment (source: Eurocontrol, 2007) 
Single dominant market 
segment 
Two dominant market segments 
Traditional 272 traditional/low-cost 41
low-cost 34 traditional/business 27
Business 24 business/other 25
Military 20 traditional/non-scheduled 23
Other 11 traditional/other 12
non-scheduled 7 low-cost/business 6
Cargo 2 business/military 6
- - traditional/military 5
- - all other combinations 8
TOTAL 370 TOTAL 153
 
The analysis of 528 airports in Europe shows that 370 airports (70%) have one 
dominant market segment2 (see Table 1 for details). At a majority of them (85%) the 
dominant market segment is one of the passenger market segments (traditional, low-cost 
or non-scheduled).  
                                                 
2 The market segment is considered as dominant if it accounts for more than 25% of the total flights. 
AIR TRAFFIC 
PASSENGER CARGO MILITARY OTHER 
SCHEDULED NON-SCHEDULED 
TRADITIONAL 
LOW COST 
CHARTER 
BUSINESS 
AVIATION 
GENERAL 
AVIATION 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
6 
 
Of these airports, 153 have two major market segments (see Table 1 for details), of 
which 42% are passenger-related, i.e. both dominant market segments are passenger 
market segments. Three dominant market segments were found at five airports only 
(two traditional/low-cost/business, two traditional/low-cost/non-schedules and one 
traditional/business other). 
Regardless of the market segment, all aircraft use the same runway system and (usually) 
taxiway system, but, if the volume of traffic justifies it, the passenger aprons can be 
separated from the cargo and general aviation aprons.  
For example, at Munich Airport (MUC), as shown in Figure 2, aprons 1, 2 and 3 are 
used for passenger traffic, aprons 8 and 9 - for cargo, aprons 10 and 11 - for general 
aviation (including maintenance). There are additional aprons - 6 and 7 (east of aprons 8 
and 9, not visible in Figure 2), located in the maintenance area. 
 
Figure 2. Aprons at Munich Airport, Germany (DFS, 2004) 
At joint civil-military airports, civil and military terminal complexes are separated. The 
taxiway systems can also be separated and used independently, if the terminal 
complexes are located on opposite sides of the runway system. The number of joint 
civil-military airports increases, as one of the solutions for the lack of runway system 
capacity in the air transport network. Instead of building new runways, the idea is to use 
nearby underutilized existing (military) runways for the overflow from congested civil 
airports.  
  
11
9
8
1 2 3 
10 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
7 
 
For example, an initiative CUMA (Civil Use of Military Aerodromes) has resulted in 
the transformation of many military airports in Europe for joint use (Siluri, 2008): 34 in 
France, nine in Spain, three in Italy, two in the Czech Republic, one in Holland, in one 
in Germany, etc. A similar initiative also exists in U.S. (MAP - Military Airport 
Programme) under which 21 military airports have been transformed into joint use by 
2008 (Siluri, 2008). 
Further on, certain segmentation can exist within the apron(s) accommodating 
passenger traffic. Significant increase in low-cost traffic caused some changes at 
airports where it became one of dominant market segments. Low-cost carriers (LCCs) 
may have their own terminal complex separated from traditional airlines. Some airports 
have built completely new terminal building to meet LCC’s requirements (e.g. 
Marseilles, Geneva, Kuala Lumpur, Singapore, Brussels-not the whole complex, but the 
new pier with 6 stands). At some airports existing terminal complexes have been 
adapted for LCCs. For example: Berlin Schoenefeld3 (the previous charter terminal has 
been adapted for the LCC Easyjet), Budapest (Terminal 1 has been reconstructed into an 
exclusive low-cost terminal, serving around 10 LCCs4), Warsaw (former Terminal 1 has 
been adapted for LCCs5), Paris Charles de Gaulle (charter Terminal 9, now Terminal 3 
is also used for low-cost flights6), Amsterdam Schiphol (one pier has been transformed 
for LCCs), (Radovanovic, 2009). 
When the share of low-cost traffic is not significant enough to justify a separate terminal 
building, LCCs prefer remote over contact stands, in order to avoid high fees for air-
bridges use. It is quite similar with the practice of non-scheduled airlines. The 
rules/constraints related to use of remote stands are not rigid usually. It is easy to 
deviate from them (which is not always the case with contact stands) if necessary to 
meet changed requirements.   
                                                 
3 Berlin Brandenburg Airport is intended to replace Schoenefeld Airport and Tegel Airport, but it has 
encountered series of delays and its opening has already been postponed on several occasions.  
4 After the bankruptcy of Malev in 2012, all airlines operating from Budapest Airport are switched to 
Terminal 2. 
5 Terminals 1 and 2 are now designated as Terminal A. South hall (former Terminal 1) is currently under 
reconstruction. 
6 EasyJet pretends to operate exclusively from Terminal 2B, after it is reconstructed. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
8 
 
1.2 Aircraft classes 
Each airport airside element has certain physical dimensions which impose constraints 
on the aircraft classes (types) that may use it.  
In the case of runways, it is their length and width that allow or not certain aircraft 
classes to use the runway. However, these constraints are not exclusive, since the larger 
aircraft can be allowed to use shorter runways by decreasing their landing/take-off 
weight. The mix of aircraft classes has a direct impact on ATC (Air Traffic Control) 
separation rules applied and related inter-event times between landings and take-offs, 
and consequently on the runway system capacity.  
At an apron, regardless of the market segment, there are physical constraints on aircraft 
stand usage due to aircraft class/stand size7 compatibility. Each stand may be occupied 
by an aircraft for which it is designed, or by any other smaller-than-design aircraft. An 
apron can consist of a fixed number of stands of a certain size or it can have a flexible 
structure that enables various arrangements of aircraft in the same area e.g. one large 
aircraft, or two smaller aircraft instead. International Air Transport Association (IATA) 
proposed, so called MARS - Multi Aircraft Ramp System concept, (IATA, 2004). The 
experience so far has indicated that the aprons with a fixed configuration are more 
sensitive to aircraft fleet mix changes. Consequently, the flexible configuration 
appeared due to rather frequent changes in fleet mix, and difficulties in coping with it.  
1.3 Airlines/alliances and flight origin/destination  
In addition to the constraints on apron usage with respect to market segments and 
physical dimensions, there are also operational constraints on terminal building usage 
that have direct consequence on apron contact stands (also referred as gates) usage.  
One of the operational constraints is caused by the necessity to separate domestic and 
international passengers, as the same processes do not apply to both categories of 
passengers, either in arrival or departure flow.  
                                                 
7 The size of the aircraft stand depends on different factors, such as: design aircraft (size and 
maneuverability), apron configuration, type of ingress and egress, clearances, ground servicing, etc. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
9 
 
International passengers have to pass through additional processes (passport control, 
customs, and perhaps additional security checks) that do not apply to domestic 
passengers. That is why terminals are divided into those handling only domestic, only 
international and mixed domestic/international passengers. This directly influences the 
allocation of the apron contact stands to particular aircraft/flight. Domestic contact 
stands can be used only by aircraft coming from and flying to domestic destinations. 
Similarly, international contact stands are available only for aircraft coming from and 
flying to international destinations. Aircraft carrying out mixed domestic/international 
flights can be accommodated at contact stands of the mixed terminals. Contact stands at 
mixed terminals can also accommodate aircraft operating any other flight 
origin/destination combination. At the San Francisco Airport (SFO), shown in Figure 3, 
contact stands A and G are international, while B, C, D, E and F are domestic. There are 
no mixed domestic/international contact stands (SFO, 2012). 
 
Figure 3. Layout of the terminal complex at San Francisco Airport, United States (SFO, 2012) 
European airports experience a somewhat more specific situation, since part of Europe 
belongs to the Schengen area inside of which a free flow of passengers and goods 
applies. Domestic passengers are those traveling not only to/from a particular country 
but to/from the entire Schengen area. Non-Schengen passengers have their origin and/or 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
10 
 
destination outside the Schengen area. Within this group it is often the case to separate 
the sub-group of international (intercontinental or special countries) passengers 
traveling to/from the countries that require some additional passenger checks (usually 
related to security). The Schengen area of the terminal building has to enable the free 
flow of passengers, and the non-Schengen area has to provide entry/exit separation from 
the Schengen area. Depending on the passenger terminal building, the apron contact 
stands can be available only for Schengen flights, only for non-Schengen/international 
flights and/or for mixed flights with respect to their origin/destination. If separation 
between non-Schengen and Schengen flows is achieved vertically (by levels) in the 
same building, contact stands can be for mixed usage. Unlike U.S. airports, there are 
many more mixed terminals at European airports, aiming to shorten connecting times 
between the flights on mixed routes that account for a great share of the total number of 
connections.  
For example, at Zurich Airport (Figure 4) Pier A is for Schengen flights, while Pier B 
and Pier E are for mixed Schengen and non-Schengen flights (vertical separation within 
the terminal). Terminal E has separate areas for international flights with special 
security requirements. Earlier (before Switzerland joined the Schengen Agreement), all 
terminals were international. Remote stands are now used for Schengen-flights 
overflow, primarily for airlines other then Star Alliance members. They are also 
engaged (in the midday peak) for Pier E long-haul flights overflow (ZRH, 2012a). 
 
B 
A 
E 
 
Figure 4. Layout of the terminal complex at Zurich Airport, Switzerland (ZRH, 2012a) 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
11 
 
At Munich Airport both terminals are for mixed flights. The separation of the flows is 
horizontal (dedicated modules) in Terminal 1, while in Terminal 2 vertical separation 
applies (dedicated levels), Figure 5. In Terminal 1 modules A and C are for Schengen, 
modules B and D for non-Schengen (B higher risk), and module E for extremely high-
risk destinations (e.g. Israel). In Terminal 2, one level for Schengen and the other for 
non-Schengen allow contact stands to be used flexibly by all flights with respect to their 
origin/destination combination8. 
 
Figure 5. Layout of the terminal complex at Munich Airport, Germany (MUC, 2013) 
If there are more dominant classes of passengers accounting for a significant share, 
different segmentation may apply, e.g. at Palma de Mallorca Airport, Spain there is an 
additional module for inter-island flights. 
Operational constraints in using passenger apron(s) may also stem from the stand 
assignment policy with respect to different airlines. In Europe, it is typical that stand 
assignment policy is entirely the matter of airport operator’s decision9, a so-called a 
common-use strategy. On the other hand, at U.S. airports, stand assignment is typically 
passed on the airlines. There is a difference between exclusive, preferential and joint use 
                                                 
8 Only for Lufthansa and Star Alliance 
9 One of the exceptions is Terminal 2 at Munich Airport, as a joint investment of the Munich Airport and 
Lufthansa, which is used only by Lufthansa and Star Alliance. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
12 
 
of stands/gates. Exclusive use signifies that a single airline has complete use and control 
of stands/gates10. Preferential use implies leasing stands/gates by a particular airline, 
while the airport operator retains the right to assign available resources to other airlines 
when they are not used by the leasing airline. Joint use assumes leasing the same stands 
by various airlines.  
For example, at John F. Kennedy Airport (JFK), shown in Figure 6, Terminal 5 is used 
exclusively by JetBlue, Terminals 2 and 3 by Delta, while other terminals (1, 4, 7 and 8) 
are jointly used by several airlines (JFK, 2012).  
 
Figure 6. Layout of the terminal complex at John F. Kennedy Airport, United States (JFK, 2012) 
At the San Francisco Airport (Figure 3) all terminals are used jointly by several 
airlines11. 
                                                 
10 Lease agreements may give the airport operator the right to negotiate for underutilized gates. 
11 Terminal 1 is used by AirTran Airways, Alaska, Delta, Frontier, Southwest and US Airways; Terminal 
2 – by American and Virgin America; Terminal 3 - by Continental, United and United Express; and the 
international area is used by all international flights, plus Hawaiian, JetBlue and Sun Country (SFO, 
2012). 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
13 
 
1.4 Impact of different user constraints on airport airside capacity 
Considering the impact on the runway system capacity (through the imposed ATC 
separation rules), the main users of the runway system are arriving and/or departing 
flights12 (regardless of airline) operated by different aircraft classes.  
On the other hand, the situation is more complex when the apron(s) are considered. In 
addition to physical constraints imposed by aircraft classes, there are other categories of 
users that can impose constraints on apron stands usage and consequently affect apron 
capacity (market segments, airlines and flight origin/destination). 
More constraints on the use of certain infrastructure element make it less flexible in 
absorbing changes in the demand characteristics. Inflexibility of one element affects the 
inflexibility of the entire system. In an inflexible system, changes in demand 
characteristics may lead to a decrease in the efficiency of overall capacity utilization.  
When an apron appears to be a bottleneck in the airport airside sub-system, it is 
necessary to detect whether there is an objective lack of capacity, or whether the 
capacity shortage appears as a result of operational imbalance between the supply 
(apron configuration and constraints applied) and the demand (demand volume and 
structure)? In the first case, the only solution to the problem is infrastructure expansion. 
In the second case, it is possible to mitigate or remove a bottleneck through adequate 
reallocation of the apron (contact and remote) stands among the users by changes and/or 
relaxation of the current constraints.  
The thesis aims in analyzing capacity of the current apron and its impact on the overall 
airside capacity at passenger airports. It also analyzes the potential to gain additional 
capacity by means of changing constraints on apron stands usage.  
                                                 
12 The runway may be designated only for arrivals, only for departures, or for mixed operations. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
14 
 
2 Airport categorizations with respect to traffic parameters 
2.1 General  
Airport categorizations are based mainly on size of airport (volume of traffic served) 
also taking into account: hubbing, presence of LCCs and, for small airports, dominance 
of regional (short-to-medium haul) traffic/routes (EC, 2008; Malighetti et al., 2009; Oetl 
and Boeck, 2012). 
Official categorizations of airports adopted by international institutions are primarily 
based on airport size, referring to traffic volume served (mainly annual number of 
passengers). In addition to that they are further classified based on dominant market 
segment(s) and traffic pattern (to split hub from non-hub type of service).  
Official categorizations (by the U.S. Federal Aviation Administration - FAA and the 
European Commission - EC) are referred to in the relevant literature on airport planning 
and design (Horonjeff et al., 2010; Ashford et al., 2011). An example is categorization 
by the Committee of the Regions, in its Outlook opinion of 2nd July 2003 on regional 
airport capacities (EC, 2005). It suggests five categories of European airports:  
1. Major Hub Airport with more than 25 million passengers per annum (4 airports, 
accounting for approximately 30% of European traffic), 
2. National Airport with 10 to 25 million passengers per annum (16 airports, 
accounting for approximately 35% of European traffic), 
3. Airport with 5 to 10 million passengers per annum (15 airports, accounting for 
approximately 14% of European traffic), 
4. Airport with 1 to 5 million passengers per annum (57 airports, accounting for 
approximately 17% of European traffic), and 
5. Airport with 200 000 to 1 million passengers per annum (67 airports, accounting 
for approximately 4% of European traffic). 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
15 
 
Some other categorizations based on number of passengers (or volume of cargo) are 
available in the sources e.g. FAA (2013), Office of the Federal Register (2011) and EC 
(2005). 
It is important to distinguish between the term “hub” that is misused in majority of 
categorization referring to airports that handle large volumes of traffic and/or to airports 
being large airline operational bases, and the real concept of a hub, referring not only to 
a large airport, but more closely associated to transfer traffic through strong 
connectivity between flights achieved by inbound and outbound flight schedule 
coordination (Dennis, 1994). Similarly, the term “national” is also misused. It is not 
used to differ between national and international, but it is again strictly connected to a 
certain volume of traffic.  
2.2 The role of airports in air transport networks  
Deregulation and liberalization in air transport had twofold effect. Traditional (also in 
referred to in literature as: full-service, ex-flag, legacy) airlines begin to concentrate 
their flights spatially and temporally at hub airports, therefore offering wider coverage 
of the routes to all “spokes” being connected to hub airport. On the other hand, low-cost 
model came into the service. In U.S. the low-cost model was established by Southwest, 
even before deregulation (in 1973) while Europe waited almost 20 years for Ryanair (in 
1995) to establish its low-cost service in the liberalized market (Dobruszkes, 2006). 
Traditional airlines organize their networks in line with the hub-and-spoke concept, 
while on the other hand LCCs operate on the point-to-point principle. 
In the hub-and-spoke network, a distinction is made between hub and non-hub airports, 
based on their role in the network. The main characteristic of hub airports is strong 
connectivity between flights. In order to achieve connectivity, one or more airlines 
concentrate their flights and operate waves (banks) of flights at hubs, enabling them to 
function as the point of exchange. They experience several (alternating 
arrival/departure) peaks of demand each day. Understanding of hub airport accounting 
for  spatial concentration and temporal coordination is now widely accepted in literature 
(Dennis, 1994; Danesi, 2006; Burghouwt and de Wit, 2005; Kraus and Koch, 2006; 
Burghouwt, 2007; etc.). 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
16 
 
Non-hub airports of the hub-and-spoke network are origin-destination (O/D) airports 
connected to hub airport(s), operating as their feeders. The term feeder is commonly 
used for these O/D airports, implying an exchange of flights/traffic in both directions, to 
and from the hub(s). In addition to feeder flights, there are other point-to-point flights 
operating from these airports. At these airports demand is not concentrated in waves, 
but spread out throughout the day, with existence of more or less pronounced peak 
periods (usually morning and late evening, and possibly a midday peak). There is no 
coordination between particular incoming and outgoing flights with the aim of 
facilitating passenger transfers at the airport. Airports connected to hub-and-spoke 
networks are usually (but not necessarily) those with predominantly traditional 
scheduled traffic. 
In addition to O/D airports operating as hub feeders in hub-and-spoke networks, some 
other airports included into point-to-point networks are also O/D airports. They mainly 
serve LCCs and/or charter airlines. Similarly as the “spoke” airports, these airports do 
not coordinate inbound and outbound flights in order to enable effective and efficient 
passenger transfers between the flights, nor they play the role of feeder airports (at least 
not as a general rule). In point-to-point network certain spatial concentration exists, 
naturally created at the strongest base airport(s) of LCCs. Usually, there is no temporal 
concentration intentionally created to facilitate transfers between flights, but it can 
consequently exist at airports with high traffic density.  
Burghouwt (2007) gives some general categorization of airports based on their temporal 
and spatial concentration, as in Table 2.  
In Table 2, airline stations refer to non-hub airports, i.e. “spokes”, traffic nodes are core 
nodes of the point-to-point networks, hubs are divided into several sub-groups (see 
following section for details) one of which is continuous (rolling) hub, which has “lost” 
wave-system structure of demand due to saturation. 
The current state of the European air transport i.e. airline/alliances networks (primarily 
hub-and-spoke and low-cost) and their foreseen evaluation are summarized in Appendix 
1. The aim is to show that, despite permanent networks evolution, airport categories 
based on spatial and temporal concentration of traffic are not likely to change. However, 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
17 
 
particular cases can undergo transformation from one role in the network to another, 
depending on the base airline strategy. 
Table 2. Airport categorization based on spatial and temporal concentration of traffic (Burghouwt, 2007) 
 Spatial configuration of the node 
Decentral Central 
Te
m
po
ra
l 
co
nf
ig
ur
at
io
n 
of
 th
e 
no
de
 No wave-system 
structure in airline 
flight schedule 
Airline station 
Feeder destination/spoke/ 
/non-hub 
Traffic node 
Continuous (rolling) hub 
Wave-system 
structure in airline 
flight schedule 
(Intermediate node) 
Hub 
Hinterland, 
Directional/Hourglass,  
All-around, Euro, 
Specialized, Global,  
Hyper, Mega, Super  
 
The role of the airport in the air transport network has significant impact on the traffic 
pattern (mainly its temporal concentration) and together with that, on utilization of the 
airport airside capacity.  
This issue is rather poorly addressed in literature, with some exceptions. Burghouwt 
(2007) devoted a chapter in the book discussing the impact of airline networks onto 
airports. He observed spatial and temporal concentration from the perspective of wave-
system structure, connectivity and seat capacity (all being certain indicators of demand 
characteristics). However, he did not address possible repercussions of the specific 
demand characteristics on airport airside capacity, which is the main issue addressed in 
this thesis.  
2.2.1 Selected airport categorizations with respect to role of airport in the 
network 
Selected airport categorizations, considering hubbing in the real sense, as summarized 
in (EC, 2008) and produced for the purpose of different studies, are given in Appendix 
2.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
18 
 
Deutche Bank Research (2006) differentiated between three categories among European 
airports (mega-hubs, secondary hubs and secondary airports).  
Buyck (EC, 2008) considered worldwide airports, and classifies them into four 
categories (primary hubs, secondary hubs, regional platforms and LCC bases).  
Burghouwt, G. (2007) differed between five categories (ranked 1st to 5th) of airports in 
hub-and-spoke networks, with respect to indirect connectivity in combination with 
presence of the wave-system structure. Although based on very important 
characteristics of hubbing, he gave the state of the network from the year 1999, which 
cannot be accepted as a good approximation of the current state due to constant 
networks evolution. For example, almost none of the categories had wave-system 
structures in 1999, which is definitely not the case at the present.  
Malighetti et al. (2009) categorized European airports accounting for volume of traffic, 
destinations of connections, connectivity and type of service. They clustered 467 
European Airports13 into eight categories (worldwide hubs, hubs, secondary gates, 
airports with 3-5 million pax, no low-cost gates, regional airports, airports for LCCs and 
local airports). However, cluster analyses should be taken with caution as they are based 
on exact performance indicators and reflect current network state, but not necessarily 
future changes. Different subgroups may result from different traffic parameters. 
Munich Airport categorization (cited in: Oettle and Boeck, 2012) defers between five 
airport categories of which two are hubs (international and secondary) and three are 
O/D airports (international, regional and secondary). 
2.3 Market segments differentiation 
Categorization of (European) airports with respect to dominant market segments is 
addressed in Chapter 1 (Section 1.1), showing the dominance of the passenger market 
segment.  
  
                                                 
13 With at least one passenger flight scheduled in autumn 2007 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
19 
 
2.4 Other considerations 
Demand for airports is rarely distributed evenly throughout the year, month and/or day. 
Instead, it is normally characterized by temporal imbalance that can be described in 
terms of hourly, daily or monthly variations in passenger, cargo or aircraft movements 
(Ashford et al, 1997; Ashford et al, 2013). Nonetheless, it is important to take into 
consideration peaking patterns on different levels (daily, weekly and yearly).  
A special sub-group of O/D airports are seasonal airports characterized by highly 
concentrated traffic during the high season, and the very low traffic during off-season 
periods. Seasonal traffic concentration is the outcome from additional demand, typically 
related to summer and winter vacation periods and related destinations, but it can also 
be imposed by events related to important holidays that may differ depending on the 
country, region, religion, etc. (Halpern, 2011). 
In Appendix 3, a list of 39 European seasonal airports is given (as of 2006; source: 
Eurocontrol, 2007). The seasonality factor is calculated as the ratio between the 
maximum number of departures during the peak day and the mean number of daily 
departures. Except “regular” seasonal airports at which seasonality is driven by the 
tourist season (summer or winter) there are several airports that have experienced 
seasonality due to special events in 2006. The majority of European seasonal airports 
experience summer peaks, in combination with common weekly peaking, so called 
“hedgehog” airports. 
Seasonal traffic patterns are illustrated on the example of several Spanish i.e. AENA’s, 
airports, given in Appendix 3.  
General principles on airside capacity analysis addressed in this thesis apply also for 
seasonal airports, but additional implications of seasonality on airport airside resource 
planning are not further elaborated.  
Airport categorization with respect to hourly and daily variations will be addressed in 
Chapter 3 (Section 3.4).  
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
20 
 
3 Airport airside capacity - definitions and factors that affect it 
3.1 Definitions of airside capacity 
Capacity is the ability of a component of the airside to accommodate aircraft. It is 
expressed in operations (arrivals and departures) per unit of time, typically in operations 
per hour (Ashford and Wright, 1992).  
In dependence on the dominant factors affecting capacity, different concepts of capacity 
may exist. Broadly, there can be operational, economic and environmental factors. They 
may work together, but in the most cases only one type of factors is dominant and thus 
determines airport airside capacity under given conditions (Janic, 2004). This thesis 
refers to operational capacity, assuming that the dominant constraining factors are 
operational.  
In general, there are two basic capacity concepts: ultimate and practical capacity. 
Saturation capacity is also known as ultimate capacity or maximum throughput 
capacity. Both capacity concepts refer to conditions of saturation (the continuous 
demand for operations is assumed) and adherence to separation requirements specified 
by the ATC. The main difference is that saturation capacity does not take into account 
the level of service (in other words it is determined regardless of delay), while practical 
capacity considers a certain acceptable level of service (specified average acceptable 
delay, usually four minutes).  
Both concepts of saturation and practical capacity are used in most relevant literature 
(Ashford and Wright, 1992; De Neufville and Odoni, 2002; FAA, 1983; Hockaday and 
Kanafani, 1974; Janic, 2009).  
As a reflection of what occurs in the case of the most congested airports in the world 
today, the runway system is considered to be the major capacity constraint. That is the 
main reason why airside capacity is usually expressed through runway system capacity. 
The definitions of ultimate and practical airport airside capacity are as it follows (De 
Neufville and Odoni, 2002): 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
21 
 
- Saturation capacity indicates the average number of operations that can be 
performed on the runway system in 1h in the presence of continuous demand, 
while adhering to all the separation requirements imposed by the ATM (Air 
Traffic Management) system.  
- Practical hourly capacity, originally proposed by the FAA in the early 1960s, is 
defined as the expected number of operations that can be performed in 1h on a 
runway system, with an average delay per operation of four minutes. The 
runway system reaches its capacity when this threshold is exceeded. The 
practical hourly capacity of a runway system is approximately 80-90% of its 
saturation capacity, depending on the specific conditions. 
These two capacity concepts are commonly used in the process of airport planning and 
development. They are delivered by all capacity estimation models for supporting 
decision making in these areas.  
It is not always suitable to express capacity in a 1-hour period, like it is typically done 
for saturation capacity or practical capacity. Other time units (smaller or larger) can, and 
probably should, be used for better description of available resources potential to meet 
particular demand requirements. This is an especially sensitive question at hub airports, 
where demand comes in waves of successive arrivals followed by a wave of successive 
departures. Hubbing parameters that may affect airside capacity are discussed in Section 
3.3, followed by the discussion about representative peak periods in Section 3.4.  
In addition to above-mentioned basic capacity concepts, there are also, sustained 
capacity and declared capacity, defined as it follows (De Neufville and Odoni, 2002): 
- Sustained capacity of a runway system is defined as the number of operations 
per hour that can be reasonably sustained over a period of several hours. 
Maximum performance often cannot be sustained in practice for a period of 
more than one or two consecutive hours.  
- Declared capacity is defined as the number of aircraft operations per hour that 
an airport can accommodate at a “reasonable” level of service (LOS). There is 
no accepted definition of “reasonable” LOS and no standard methodology for 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
22 
 
setting it14. The declared capacity is roughly 85-90% of the saturation capacity 
of the runway system15. 
These two concepts are meant rather for describing reasonable capabilities of existing 
systems. Declared capacity is used in specifying the number of slots available for 
schedule coordination purposes, based on specific LOS that airport aims to meet. In 
addition, it is widely used for airport benchmarking, although many different 
understandings of this capacity concept can leads to inconsistency in comparison.  
In addition to capacity in absolute terms, its utilization has shown to be particularly 
important performance indicator, not only for peak periods, but also for longer time 
horizons (day, week and season). The level of capacity utilization can indicate 
underutilization of the capacity (as a consequence of traffic demand being lower than 
capacity), or inefficiency in capacity management, particularly at congested airports 
(when delays occur due to inability to accept all scheduled operations on time). 
Cherniavsky and Abrahamsen (2000) defined airport utilization metrics (as a subset of 
FAA Aviation System Performance Metrics) and explained an approach to calculate 
them. In brief, arrival/departure/airport utilization represents a degree to which 
arrival/departure/mixed demand was satisfied for a given period, taking into account the 
airport’s target capacity in that time period. These measurements taken over 15-minute 
periods are then combined to produce utilization performance (arrival, departure and 
overall) for the day.   
3.2 Factors that affect airside capacity 
Different approaches to classifying factors affecting airside capacity have been used in 
the literature (Ashford and Wright, 1992; De Neufville and Odoni, 2002; Janic, 2009; 
Newell, 1979).  
Based on research and practice, major factors affecting runway system capacity are 
classified in four groups in this thesis. The factors belonging to each group are listed 
below.  
                                                 
14 Such as four minutes typicaly used for practical capacity 
15 The declared capacity is not  necessarily imposed by the runway system LOS. It can also be affected by 
the LOS specified for apron(s) and terminal building.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
23 
 
1. Runway system layout – number of runways, runway length and runways’ 
mutual position (parallel, crossing, extended runway centerlines crossing (V 
configuration)), number of simultaneously active runways, available runway 
configuration with respect to type of operations (departures/arrivals/mixed 
operations), number and location of runway exits, existence and position of 
crossing taxiways; 
2. Demand characteristics – traffic distribution, exchange of operations 
(arrivals/departures), fleet mix, mix of market segments (scheduled, low-cost, 
charter, cargo, business, etc.); 
3. Operational constraints – separation rules set by ATC requirements, runway 
occupancy times;  
4. Local conditions – meteorological conditions, environmental conditions. 
These four groups of factors are not strictly separated, but they overlap to some extent. 
For example, runway exit location is an infrastructural factor, but at the same time it is 
also an operational factor, because it directly affects runway occupancy time, which 
then affects arrival-arrival and arrival-departure separation at a particular airport; or 
local meteorological conditions, especially wind and visibility, have direct impact on 
configuration of the runway system in use.  
Although the runway system is identified as the major airport scarce resource, the 
capacity of other infrastructure elements should also be taken into account.  
The taxiway system at major airports is most often designed to provide capacity which 
exceeds the capacity of the runway system and thus not be the factor limiting airside 
capacity (De Neufville and Odoni, 2002). This does not necessarily mean that some 
local constraints cannot be identified, especially in the areas of taxiway intersections, 
points where taxiways cross an active runway(s) or where high-speed exits merge with 
taxiways. The general rule is that taxiway capacity problems are specific for each 
airport and must be resolved at the local level, for the specific configuration and under 
the local conditions. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
24 
 
On the other hand, apron capacity can be a limiting factor of the overall airside capacity. 
Airports have lately been facing this more often. Apron capacity depends on similar 
factors as the runway system capacity. The factors can be classified in four groups, 
similarly as those for the runway system: 
1. Apron layout – number of parking stands, number of stands per market segment 
and per aircraft class, number of stands that can be simultaneously used with 
respect to lateral separation, apron taxiway layout; 
2. Demand characteristics – traffic distribution, exchange of operations 
(arrivals/departures), fleet mix, mix of users (classified by different criteria - 
airlines, or flights O/D combination), mix of market segments (scheduled, low-
cost, charter, cargo, business, etc.); 
3. Operational constraints – turnaround times, buffer times between aircraft which 
use the same or laterally endangered stand, policy of parking stand usage (e.g. 
by airline in U.S., or by flight O/D in Europe); 
4. Local conditions – meteorological conditions (particularly visibility). 
The factors that affect runway system and apron capacity are summarized in Table 3. 
Although capacities of the runway system and apron depend on similar factors (see 
Table 3), it does not imply that both of them reach their capacities under the same 
conditions. By assessing capacity of the runway system and capacity of the apron 
complex we have a certain, but rather incomplete, picture of the overall airside capacity 
and its sensitivity on demand changes. It is necessary to consider their functional 
relationship in order to create the overall airside capacity scheme. 
The thesis places a special emphasis on apron capacity, which is modestly addressed in 
academic and professional literature, including the explicit consideration of the 
functional relationship to the runway system and the differences induced by demand 
characteristics, particularly traffic distribution depending on the role of the airport in air 
transport networks (O/D airports vs. hubs). 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
25 
 
The characteristics of hub airports with implication on airside capacity utilization are 
discussed in the reminder of this chapter (Section 3.3). 
Table 3. Factors that affect runway system and apron capacity 
Factors that 
affect capacity RUNWAY SYSTEM APRON 
Design 
parameters 
Runway system layout (number of 
runways, runway lengths and 
runways’ mutual position: parallel, 
diverging/converging, crossing);  
Runway system configurations 
(number of simultaneously active 
runways, runway allocation with 
respect to type of operations: 
departures/ arrivals/ mixed 
operations);  
Number and location of runway 
exits; existence and location of 
crossing taxiways 
Number of parking stands;  
Apron layout (number of stands per 
market segment, and per aircraft 
class);  
Available apron configurations in 
terms of number of stands that can be 
simultaneously used with respect to 
lateral separation (for aprons with 
MARS stands);  
Apron taxiway system layout; 
Demand 
characteristics 
Traffic distribution and exchange of 
operations (arrivals/departures);  
Fleet mix; 
Mix of market segments 
(scheduled, charter, cargo, business, 
etc.) 
Traffic distribution and exchange of 
operations (arrivals/departures);  
Fleet mix;  
Mix of users (classified by different 
criteria - airline, or type of service);  
Mix of market segments (scheduled, 
charter, cargo, business, etc.) 
Operational 
considerations 
Separation  rules set by ATC 
requirements;  
Runway occupancy times 
Turnaround times;  
Buffer times between aircraft using 
the same or laterally endangered 
stands;  
Policy of stands usage  
Local 
conditions 
Meteorological conditions 
(visibility, ceiling, wind); 
Environmental conditions 
(emissions, noise) 
Meteorological conditions 
(particularly visibility) 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
26 
 
3.3 Characteristics of hub airports with implication on airport airside capacity 
3.3.1 Spatial and temporal concentration at hub airports 
At hub airports, the core nodes of the hub-and-spoke networks, the dominant airlines 
concentrate their flights and operate waves (banks) of flights (Figure 7) in order to 
achieve strong connectivity between flights.  
 
Figure 7. A wave of flights at a hub airport (Dennis, 1994) 
The main characteristics of hub-and-spoke networks are their spatial and temporal 
concentration.  
Spatial concentration can be defined by the level of airline network concentration 
around one or a few hub airports (Burghouwt and de Wit, 2005). It is mainly determined 
by the geographical location of the hub airport, especially in Europe. For example the 
Paris and Brussels area is favorable for intra-European connections, while for long-haul 
connections the entire area between London and Zurich may be considered as 
competitive (Dennis, 2005). In other markets local demand may additionally induce 
spatial concentration. 
Temporal concentration assumes coordination between inbound and outbound flight 
schedules aimed at achieving higher number and quality of indirect connections. 
Demand at hub airports comes in a wave of arriving flights, followed by a wave of 
departing flights, allowing sufficient time for transfers between them. Usually, several 
waves are scheduled during a day.   
Out 
In 
Hub  Time 11:30 11:0010:3010:00
DEPARTURES 
ARRIVALS 
MINIMUM 
CONNECT TIME 
source: Dennis (1994)
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
27 
 
Another vital requirement for effective airline hub operation is sufficient airport airside 
and landside capacity, to handle a large number of aircraft simultaneously and process 
transfer passengers in a short period of time, providing some margin to absorb delays 
(Dennis, 1994). 
3.3.2 Ideal wave 
An ideal wave assumes a complex of incoming and outgoing flights, structured such 
that all incoming flights connect to all outgoing flights. The wave starts with the arrival 
time window, followed by transfer period and completed with departure time window 
(Danesi, 2006).  
The structure of the wave is determined by:  
1. The minimum connecting time – MCT, which depends on the category of 
connections),  
2. The maximum acceptable connecting time – MaxCT, and  
3. The maximum number of flights that can be scheduled per time period - N.  
The structure of an ideal wave containing intercontinental and continental connections 
is shown in Figure 8.  
 
Figure 8.  Ideal wave (Bootsma, 1997, cited in: Burghouwt, 2007) 
In this case, time windows for arriving intercontinental (ICA) and continental (EUR) 
flights and departing continental and intercontinental flights, are defined as (Burghouwt, 
2007): 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
28 
 
‐ ICA arriving window: [C-Ti+0.5Tc, C-Mi+0.5Mc],  
‐ ICA departing window: [C+Mi-0.5Mc, C+Ti-0.5Tc],  
‐ EUR arriving window: [C-0.5Tc, C-0.5Mc], 
‐ EUR departing window: [C+0.5Mc, C+0.5Tc], 
where: 
C – the wave center (point 5, Figure 8), 
Tc – the MaxCT for continental connections (2-8, Figure 8), 
Ti/c –the MaxCT involving intercontinental connections (1-8, Figure 8),   
Mc – the MCT for continental connections (4-6, Figure 8), and 
Mi/c – the MCT involving intercontinental connections (3-6, Figure 8). 
 
Connections have to meet MCT in order to be viable. MCT depends on the airline, type 
of connection (domestic, continental, intercontinental, or other) and airport design and 
performance (i.e. capacity to process transfer passengers and baggage within and 
between terminals). Typical MCTs at world airports are given in Appendix 4. European 
average is clustered around 45min, with small differences between airports (Dennis, 
1994). Exceptions are modern terminals (purpose-built) where these times are decreased 
to 35min (Munich Airport) or even 25min (Vienna Airport). Connections to long-haul 
destination, involving changing terminals, require longer times (75min Terminal 1 to 
Terminal 4 at London Heathrow).  
MaxCT reflect the LOS thresholds that keep connections attractive to passengers. It 
depends on the type of connection. For example, long-haul passengers are generally 
ready to wait longer than short-haul passengers. Bootsma (1997) (cited in: Danesi 
(2006)) has defined MaxCT for different types of flights and different quality 
thresholds. For poor quality level MaxCT is considered to be 180min for continental 
connections, 300min for continental-intercontinental connections and 720min for 
intercontinental connections.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
29 
 
A trade-off has to be made between the maximum acceptable connecting time and the 
maximum number of flights that can be scheduled in arrival/departure time windows 
(Burghouwt and de Wit, 2006). The one that is more critical, determines the size 
(duration) of the wave. In other words, the size of the wave depends mainly on the 
runway system capacity, apron capacity and on the LOS threshold (i.e. MaxCT). 
As a reflection of the practice at major European hubs, the typical wave size is around 
50-60 aircraft, with an arrival and departure time-wave of approximately one hour for 
each. In typical U.S. wave 60 aircraft arrive in 30 minutes interval, and after being on 
the ground for at least 30 minutes, they all depart within 30 minutes (Dennis, 2005). 
Ideal wave is unlikely to exist in practice due to different reasons (Burghouwt and de 
Wit, 2005). Some of them are that the “spokes” are too close, or too far, to fit in the 
wave structure, strong O/D market that attracts flying off-wave, fleet utilization or other 
reasons airlines choose not to adopt wave-system structure, and last but not least, 
capacity constraints. 
Concentration of flights in waves affects turnaround times and consequently apron 
capacity. Furthermore, underutilized periods between consecutive waves affect 
efficiency of airport airside capacity utilization. 
For example, if the runway system capacity is 60 arrivals or departures per hour and 
MCT is 30 minutes, the most of 90-minutes connections (MaxCT=90min) are 
achievable with 30 flights/aircraft per wave. The average turnaround time increases at a 
linear rate, and reaches 60 minutes for a 30-flights wave and 90 minutes for a 60-flights 
wave (Dennis, 2005). This has a direct impact on apron capacity.  
3.3.3 Hubbing performance 
When it comes to indication of hubbing performance, i.e. an influence of temporal 
concentration on airport performance, indirect connectivity is often addressed in the 
literature as a global measure of temporal concentration at hub airports.  
Different approaches are used to evaluate indirect connectivity. They are all based on 
the analysis of the number and quality of connections, but the criteria that define quality 
of connections differ between them.   
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
30 
 
For example, Burghouwt and de Wit (2005) calculate weighted indirect connectivity 
index as the sum of weighted indirect connections at an airline hub. For all possible 
connections between arriving and departing flights during the same day, they calculate 
the weighted indirect connection number (WNX) as a function of the quality of the 
connection at the hub airport (calculated based on: MCT, MaxCT and transfer time) and 
the quality of the indirect flight compared to direct flight (ratio of the actual indirect 
flight time and estimated direct flight time). Hub connectivity is judged as high if WNX 
is over 2500, medium if it is between 500 and 2500 and low if less than 500.  
Some other indirect connectivity measures are briefly described in Appendix 5.  
This thesis does not observe temporal concentration of flights in the context of indirect 
connectivity, but it rather focuses on implications of temporal concentration of flights 
on airside capacity.  
3.3.4 Wave-system structure 
Wave-system structure is specified by (Danesi, 2006, Burghouwt and de Wit, 2005): 
1. Number of waves,  
2. Wave repeat cycle (WRC) - the time interval between the same points of the 
consecutive waves, and  
3. Wave-system structure which may be broken or complete, depending on where 
short and medium haul fleet is located overnight, at hubs, spokes or both.  
Analysis of the European hub airports since 1990s shows changes of the traffic patterns 
at hub airports from rather “uniform”, i.e. spread out throughout a day with more or less 
pronounced peaks, to wave-system structure (Borghouwt, 2007). An example of 
Munich Airport is given in Appendix 6.  
Burghouwt (2007) has used a theoretical approach16 to analyze wave-system structures 
at main European hubs, i.e. to identify its presence, the number of waves, and the timing 
                                                 
16 Explanation is given in Appendix 6. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
31 
 
of the waves17. However, his approach has omitted to catch an exchange between 
arrivals and departures, which is rather important from the perspective of airside 
capacity utilization efficiency. 
An exchange of arriving and departing segments of the wave-system is clearly visible 
on real traffic data. It can be more or less pronounced depending on the type of hub 
airport, i.e. of the maturity level it approached in its transformation from an O/D airport.  
The wave-system structure is much more obvious at small airports which try to improve 
connectivity between flights through better coordination between inbound and outbound 
flights (Danesi, 2006). Figure 9 gives an example of clear wave-system structure18 at 
Milan Malpensa Airport (before withdrawal of Alitalia).  
 
Figure 9. Daily traffic at Milano Malpensa, Italy, 19th January 2005 (Danesi, 2006) 
However, at some large hubs, the wave-system structure is difficult to recognize, as is 
the case at London Heathrow, see Figure 10.  
                                                 
17 Some examples of airports with good wave-system structures (at that time, year 2003) are: Amsterdam 
- KLM operated 5 waves; Frankfurt - Lufthansa operates 5 waves; Munich - Lufthansa operated 4 waves; 
Zurich - Swiss operated 6-7 waves and Vienna - Austrian operated 5 waves. 
18 It is described as (3,51/2, 1) standing for three waves with repeat cycle on 5.5h and fleet stabled at 
“spokes” (Danesi, 2006). 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
32 
 
 
Figure 10. Daily traffic at London Heathrow, United Kingdom, 19th January 2005 (Danesi, 2006) 
3.3.5 Evolutionary phases 
Kraus and Koch (2006) argue that strict separation between O/D and hub airports does 
not reflect the full range of airport roles in air transport networks. They analyzed 
arriving and departing seat capacity both for the entire airport and for its hub airline, or 
the strongest airline at the airport19. They identified five different evolutionary stages 
i.e. types of airports: 
1. O/D airports (without hub characteristics), e.g. Tunis (Figure 11), 
2. O/D airports with first hub characteristics, e.g. Johannesburg (Figure 12), 
3. Long-haul transfer hub airports, e.g. Doha (Figure 13), 
4. Mature hub airports, e.g. Paris Charles De Gaulle (Figure 14), and 
5. Hub airports with de-peaking strategy (rolling hubs, continuous hubs), e.g. 
Dallas Fort Worth (Figure 15). 
At type 1 airports, the seat capacities offered are in line with the demand. Several peaks 
may occur during the day. At Tunis Airport (Figure 11) morning departure peak results 
from the airline strategy to base almost its entire fleet at home base overnight. Midday 
                                                 
19 The entire flight schedule data (for calendar week 10 in 2006), including departure and arrival times, 
offered capacity per flight and operating airlines, has been analyzed and clustered into 48 time periods of 
30 minutes. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
33 
 
peak results from the airline strategy to serve destinations reachable in 2.5 hours flying 
time implying that aircraft return to the base airport within approximately the same 
time-frame. The second peak is created when aircraft leave for their second daily 
rotations. 
 
Figure 11. Type 1, O/D airport: Tunis Airport, Tunis (Kraus and Koch, 2006) 
At type 2 airports (Figure 12) demand is characterized by a limited overall share of 
home base airline, as a result of the competition by other airlines in the liberalized 
market with still strong O/D market. In such case, home base airlines concentrate their 
flights in one or two waves per day. 
 
Figure 12. Type 2, O/D airport with first hub characteristics: Johannesburg  Airport, South Africa (Kraus 
and Koch, 2006) 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
34 
 
At type 3 airports (Figure 13) wave-system structure is clearly visible. Long-haul routes 
are served at a lower frequency than short- and medium-haul routes, thus requiring 
longer times for passenger transfers. This results in a smaller number of “wider” waves. 
Airport infrastructure is highly utilized during peak periods, while during off-peak 
periods it remains underutilized. 
 
Figure 13. Type 3, long-haul hub: Doha Airport, Qatar (Kraus and Koch, 2006) 
At well-established hub airports, type 4 (Figure 14), strong home base airline clearly 
dominates, operating several waves (5-6) during the day, each with a 1.5-2.0 hours 
duration. At these airports waves follow each other at short intervals (towards their 
overlapping), which leads to more or less stable utilization of airport infrastructure. 
Waves are interrupted by short periods of lower traffic loads, which are sometimes 
useful for recovery after disturbances. 
At type 5 airports (Figure 15) the home base airline dominates. Due to the increase in 
traffic, not followed by an increase in capacity accordingly, the system cannot continue 
to function efficiently by keeping a clear wave-system structure, allowing for 
underutilized periods (idle slacks between waves). In order to increase the overall 
utilization of airside capacity through its permanent utilization, the traffic pattern 
changes accounting for larger number of shorter waves that follow each other in shorter 
time periods, leading to their overlapping after a further increase in the number of 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
35 
 
waves. So, over the time wave-system structure disappears, and the traffic pattern 
appears to be rather “uniform” at these airports.  
 
 
Figure 14. Type 4, mature hub: Paris Charles de Gaulle Airport, France (Kraus and Koch, 2006) 
  
 
Figure 15. Type 5, rolling (continuous) hub: Dallas Fort Worth Airport, United States (source: Kraus and 
Koch, 2006) 
This process of “destroying” the wave-system and returning to “uniform” pattern during 
the day is known as continuous or rolling hubbing. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
36 
 
It is not a new concept, but it preceded when connections were not created “artificially”, 
but naturally, as a consequence of high volume of flight activity at busy airports 
(Dennis, 2005). By keeping the aircraft moving all day instead of concentrating them 
into fixed waves, it certainly increases utilization and productivity of the airport. On the 
other hand, it decreases connecting opportunities by offering a lower number of quality 
connections. Due to that, this may be a “solution” for LCCs, as the difference in price 
they offer may stimulate passengers to accept longer waiting times between connecting 
flights (Dennis, 2005). 
3.4 Peak period 
Airport airside capacity analysis and planning can be based on annual, daily and hourly 
traffic volumes and structures20. Which traffic data should be used depends on the type 
of question asked and the time horizon (planning level) considered.  
Annual figures21 are typically used to illustrate (describe) overall airport performance, 
and to draw a rough conclusion about whether the airport is capable of handling 
forecasted traffic volume(s) in particular year(s) in the future, usually in a time-frame of 
15-20 years ahead (de Neufville and Odoni, 2002).  
For a more detailed analysis daily and hourly traffic figures are far more important. 
Depending on the type of capacity analysis, peak day and peak hour also referred to as 
design or typical day/hour, are commonly used. Runway system capacity analysis is 
typically based on peak-hour traffic. On the other hand, determining whether available 
contact and remote stands have adequate capacity to accept demand requirements, asks 
for working with the design day.  
Peak hour is not the busiest hour of the year, but it is a busy hour that is exceeded for an 
acceptably small number of hours during the year. Depending on the country or region, 
                                                 
20 Passenger figures are commonly used for landside planning and design, while for the purpose of airside 
planning and design it is air traffic operations that are of importance. 
21 They are commonly expressed in number of passengers, number of aircraft operations and a cargo 
volume. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
37 
 
different approaches to defining peak hour apply. The most common ones are given in 
Appendix 7 (de Neufville and Odoni, 2002).  
Similarly the design day is not the busiest day of the year, but it is exceeded for an 
acceptably small number of days during the year. Common practices in determining 
peak day volume are given in Appendix 7 (de Neufville and Odoni, 2002). 
For the analysis of the available resources utilization it is important to take into 
consideration arrival and departure distribution in time, as well as changes in traffic 
structure throughout the day.  The Transportation Research Board (TRB) Airport Coo-
perative Research Program (ACRP) report (TRB, 2012a) on “Preparing peak period and 
operational profiles – guidebook”, differs between design day profiles that assume 
arrival and departure distributions, and design day schedule that contain flight-by-flight 
information on airline, aircraft type, flight time, origin or destination, etc. The format of 
the schedules depends on their intended use. Design day schedules are necessary for the 
analysis on the highest level of detail (airfield simulations). For the purpose of tactical 
and strategic planning airside capacity analysis does not necessarily require detailed 
design daily schedules, as they are usually performed throughout less detailed models, 
thus requiring less detailed input (arrival rates, fleet mix, user mix for typical peak 
periods, but also certain schedule parameters, such as wave-system characteristics at 
hub airports). 
In some cases, instead of using peak hour and design day, a more convenient approach 
would be identification of representative peak periods that best describe design day 
traffic. Namely, a single peak period or design day may not catch fluctuations in traffic 
structures during the day or week (e.g. waves centered on long-haul intercontinental 
connections vs. waves centered on medium-haul continental connections or short-haul 
regional connections) and traffic patterns during the week or season (e.g. only jumbo-
jets during Hajj period vs. mixed fleet during off-Hajj periods at Medina Airport). 
Depending on the airport, daily traffic patterns may have different number, duration and 
level of peak periods. Thus, design day may be described with one or more 
representative peak periods. This applies both to the runway system and apron capacity 
modeling, assessment and analysis.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
38 
 
Ashford et al. (2013) discuss the nature of peaking with respect to the most important 
factors affecting the distribution of peaks throughout the day: domestic/international 
ratio; charter and low-cost/scheduled ratio; long-haul/short-haul ratio; geographical 
location and nature of catchment area. Although the timing of the peaks is important for 
capacity analysis it is also important to know the structure of the peaks. Parameters 
describing peak periods are their duration and traffic structure with respect to aircraft 
types (for runway system) and both aircraft types and airlines/other users (for apron 
capacity analysis).  
Oettl and Reeb (2012) suggested an approach to categorizing airports based on traffic 
parameters, mainly for the purpose of the runway system related capacity evaluations. 
An overview of the approach to deriving representative airport categories based on 
traffic parameters is shown in Figure 16. 
 
 
Figure 16. An approach to identifying airport types based on traffic parameters (Oettl and Reeb, 2012) 
Main traffic parameters for airport categorization were adopted to be traffic peaking and 
related parameters. Each traffic peak is described with its aircraft type structure (with 
respect to 10 aircraft weight classes) and peak shape related parameters: peak duration 
in hours, peak fill factor (as ration between area under the peak and area of the 
rectangle) and peak amplitude (as percentage of the maximum peak at the respective 
airport), as it is illustrated in Figure 17.  
Appendix 8 shows 11,936 traffic peak situations clustered in 19 representative traffic 
peak situations.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
39 
 
 
Figure 17. Traffic parameters (Oettl and Reeb, 2012) 
Apart from parameters that characterize traffic peak situations, authors also categorized 
airports with respect to parameters describing daily traffic. These are: distribution of 
traffic throughout the day and daily traffic mix. Parameters describing distribution of 
traffic during the day are: number of peaks, number of arrival and departure peaks, fill 
factor (as ratio between the area under the traffic distribution graph between 7h and 23h 
and the area of the rectangle given by the maximum number of operations during the 
day), relative load (as ratio between the time period of flight activities at or above 80% 
of the maximum number of operations and the time period of flight activities at or 
above 20% of the maximum number of operations) and relative night rest (as ratio 
between total time period with number of operations below 5 in 30min period and 8h 
default nighttime period 23h to 7h). Figure 18 illustrates fill factor and relative load. 
 
Figure 18. Fill factor and relative load (Oettl and Reeb, 2012) 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
40 
 
Cluster analysis of 203 world airports22 with respect to traffic peak parameters resulted 
in 16 representative airport categories. They are given in Appendix 8. 
It is important to notice that duration of peak periods ranges between 2.5h and 3.5h with 
only two exceptions over 5h. This confirms the above-mentioned discussion about using 
representative traffic peak periods instead peak hours. It is supported by Table 4 which 
gives the typical representatives for all 16 categories of airports. Each airport is 
represented with three main peak periods, e.g. airport category 11 has first 
representative peak type 9, second peak type 3 and third peak type 10.  
Table 4. The first thee representative peaks for each airport category (Oettl and Reeb, 2012) 
 
Airport categorization based on traffic parameters, as explained above, does not 
consider exchange/mix of arrivals and departures in peak periods. The number of total, 
arrival and departure peaks result from the separate analysis of total, arrival and 
departure traffic, respectively. With this approach information about slack between 
arrival and departure peaks (important for hub airport’s capacity analysis), or share of 
each in overlapped peaks (at O/D airports and saturated hubs) is not captured. However, 
this would be another traffic parameter useful for the analysis of capacity utilization 
both of the runway system and of the apron(s). 
Furthermore, this airports and peak periods categorization does not consider the 
characteristics of airlines/other users. The criteria for “users” depend on the region in 
which airport is located, its role in air transport network, dominant market segments at 
the airport, etc. (e.g. airlines, O/D, type of service, etc.). Due to that, it is not easy to 
categorize airports with respect to user mixes as an additional traffic parameter.  
Oettl and Bock (2012) addressed the issue of airport categorization with respect to other 
traffic parameters (in addition to aircraft fleet mix), such as: hubbing characteristics, 
flight distances and share of airlines. They used the example of secondary European 
                                                 
22 With more than 18 operations per hour in the dataset analyzed (June 2008) 
Representative 
airport 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1st relevant peak 1 2 1 1 1 9 1 2 6 1 9 11 10 12 17 18
2nd relevant peak 4 5 2 6 2 10 2 5 8 13 3 14 11 13 16 19
3rd relevant peak 7 7 3 8 4 3 7 1 7 3 10 6 13 15 9 14
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
41 
 
hubs to create representative scenarios. The results are given in Appendix 8. All 
parameters were observed independently of each other, which caused the loss of 
important information about traffic characteristics. The authors themselves argue that 
this approach is not promising. The differences may be quite significant, even in the 
similar group of airports, and thus difficult to generalize.  
The bottom line is that it is difficult to generalize representative peak periods for apron 
capacity analysis. They should be defined on case-to-case basis. 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
42 
 
4 Airport airside modeling - literature review 
A broader literature overview of the airside modeling is given in Mirkovic (2010). 
Updated findings are summarized in this chapter.  
The first part provides an overview of the models for airside capacity assessments and 
analysis. It addresses modeling of the runway system, taxiway system, apron(s) and the 
entire airside, which are mainly used for supporting airport planning in the medium- and 
long-term. Broader literature overview has shown that the most mature areas are runway 
system modeling and simulation modeling of the entire airside. 
Additionally, an overview of the models/tools for optimizing available resource 
utilization is given in Appendix 9. These models and tools are primarily designed for 
and implemented at major airports, aiming to support airside short-term decision 
making (capacity management). The overview of the literature has shown that the 
overall emphasis is again on optimization of the runway system utilization, regardless of 
whether they treat runway system solely, or as a part of the entire airside. Regarding the 
apron complex, the gate assignment problem is widely addressed in the literature. 
However it considers gates mainly in the context of the landside. 
Second part of this chapter addresses the issue of apron capacity modeling, which is not 
often referred to in literature. Apron is considered as a component of the airside (but 
still not separating it from the operational constraints imposed by the landside). Special 
emphasis is placed on apron analytical modeling. 
4.1 Airport airside capacity models for supporting planning studies 
The history of analytical (macroscopic) runway system modeling started in 1959, by 
Blumstein. Blumstein (1959) set the method for calculating landing capacity for a single 
runway. Since Harris (1972), many extensions and improvements of the Blumstein’s 
method have been developed. They have included capacity of mixed operations, 
capacity of different runway system configuration, introduction of new technologies and 
procedures, etc. One of the first was Tosic and Horonjeff (1976). They analyzed the 
effects of the microwave landing system (MLS) introduction on runway system landing 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
43 
 
capacity. One of the latest papers on runway system analytical modeling is: Janic 
(2008). Janic (2008) developed analytical models for calculating the ultimate arrival, 
departure, and mixed operation capacity of closely-spaced parallel runways using 
innovative approach procedures - staggered approach procedure (SGAP) and the steeper 
approach procedure (SEAP).  
The basic runway system capacity estimation method set by Blumstein and its 
extensions served as the basis for FAA Airfield Capacity Model (Swedish, 1981) and 
LMI Capacity and Delays Model, which represent the same method translated into 
computer language, enabling faster calculation and thus being more convenient for the 
analysis of different scenarios. Modern, improved versions in this category are: Airport 
Capacity Analysis Through Simulation (ACATS), described by Barrer and Kuzminski 
(2005) and runwaySimulator, by Barrer (2007). The advantage of the latest models is 
their flexibility. They can be easily applied to any airport in the world (any 
configuration and set of separation rules) without changing the simulation code itself. 
Regarding taxiway and apron modeling, not much was done in the field of macroscopic 
modeling. (Detailed literature review on apron modeling in given in the next section.) 
Although they depend on almost the same factors as the runway system, the operational 
constraints and relations to other airside elements are very locally specific and very 
difficult to generalize and observe isolated from the entire airside system, as can be 
done for the runway system. Therefore aprons are usually observed, modeled and 
resolved on a case-by-case basis and they usually exist as one of the modules of 
integrated high- level-of-detail simulation models. 
The most common simulation tools being used nowadays for airside analysis are: 
TAAM23, SIMMOD24Plus/Pro and RAMS25Plus. All three are complete gate-to-gate 
models, while some others, such as: The Airport Machine and HERMES26 (developed 
for Heathrow and Gatwick Airports) are limited only to the airport airside area. A 
detailed description and evaluation of these models can be found in: Odoni and Simpson 
(1979); Odoni (1991), Gass (1992); Odoni et al. (1997), and up-to-date coverage at 
                                                 
23 Total Airspace and Airport Modeler 
24 Airport and Aispace SIMulation MODeler 
25 Reorganized ATC Mathematical Simulator 
26 Heuristic Runway Movement Event Simulation 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
44 
 
Ashford et al. (2011). Updates of new features are available on the websites of 
companies currently in charge of the development and distribution of the tools: TAAM 
– Jeppesen, (2013), SIMMOD – ATAC (2013) and for RAMSPlus – ISASoftware 
(2013). AirTOpsoft (2013) represents a new generation of gate-to-gate fast-time 
simulation tools. It does not introduce much innovation in modeling the airside itself, 
but has the advantage of integrating future or customer-specific ATC concepts much 
faster. IATA’s Total AirportSim includes terminal buildings within its scope in addition 
to airspace, runway system, aprons and gates (IATA, 2008). 
Fast-time simulation models are of significant importance to airport planning. They 
enable definition of the system in high detail, by including physical and operational 
constraints that apply at the airport. Once the model is built, which requires a major 
effort, one can determine the effects caused by demand changes (demand increase, 
demand characteristics changes, temporary changes caused by the occurrence of certain 
events, etc.), by changes on the supply side (new runway, new parking stands, new 
runway exits, closure of a runway or an apron due to reconstruction, etc.), or by 
operational changes (procedural or technological improvements, new concepts 
introduction, etc.) in a relatively quick and reliable way, which makes it possible to 
analyze different scenarios and to choose the optimal solution based on that. 
Simulation models are designed to deliver highly detailed output data, which is very 
useful and required in some stages of planning, but not necessary every time, especially 
in the case of strategic planning. Although capable of delivering a high level of output, 
it is important to remember that the quality of output depends on the quality (amount 
and level of detail) of input data provided. The more uncertain the input data are, the 
less we can trust the delivered output data. It is often the case that there is a lack of 
available input data for feeding into the simulation. In such cases, simulations are not 
the solution that decision makers should reach for and rely on. It is something that is 
often neglected by airport planners, and the reason why these tools are sometimes 
misused. 
Due to that, when we have to calculate with rough input data macroscopic models are 
more convenient. Dealing with the uncertain future, planners, managers and designers 
need to recognize the wide range of situations that may occur, examine the implications 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
45 
 
of these scenarios, and develop strategies that enable them to seize opportunities and 
protect them from risk (TRB, 1987). In such situations we can gain more from 
macroscopic models. An advantage of macroscopic models, in comparison to simulation 
models, is their capability to deliver good enough output very quickly, with less detailed 
input that is usually all that is available to the planner (e.g. arrival rates instead of the 
flight schedule). “Quickly” does not refer to calculation time, which is also quite small 
in simulations, but to the preparation period which is much more time-consuming for 
simulation modeling. That makes them more suitable for analysis of a great number of 
scenarios and allows their relative comparison. Such models are not necessarily used as 
a support to final decision-making. It is desirable to use them for the selection of 
candidate scenarios that will further be analyzed through high-level-of-detail simulation 
tools. 
One of the rare integrated airside macroscopic models is MACAD - Mantea Airfield 
Capacity and Delay. It was introduced by: Andreatta et al. (1999); further elaborated in: 
Stomatopulos et al. (2004); suggested as a component of the advanced decision support 
tool for total airport performance assessment and capacity management by EC (2006). 
MACAD integrates macroscopic airside models to provide approximate estimates of the 
capacity, utilization and delays associated with every element of the airside. It is 
primarily meant for studies that require only approximate answers while examining a 
wide range of hypotheses and scenarios regarding future conditions at an airport. 
MACAD obtains reliable approximations quickly, even with a limited set of inputs 
(which is usually all we can provide/obtain for the analysis). It consists of five modules: 
airside, weather, detailed schedule generation, coordination and user interface. The 
airside module consists of a runway system capacity model (a generalized stochastic 
analytical model for the estimation of the capacity envelope of the runway system, for a 
wide variety of runway system configurations), a runway system delay model (an 
analytical model for computing the distributions of delays throughout the time interval 
of interest), and an apron/taxiway macroscopic simulation model (in which taxiways are 
represented by probability distributions for the taxi-in and taxi-out times for each 
configuration of the runway system). There are no specific details given on apron 
modeling, but it is stated that the this module identifies the stands which are most 
limiting (depending on the aircraft types, the type of flights, and the handler/airline that 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
46 
 
they serve) as well as the ones that are underutilized for the examined configuration and 
demand scenarios. 
In 2009, the project “Evaluating Airfield Capacity” was initiated under the TRB -
ACRP. The aim was to produce a new manual for capacity analysis intended for airport 
planners. The report (TRB, 2012b) was released in December 2012. It provides another 
software solution (spreadsheet) for runway system capacity estimation, but does not 
suggest any further improvements in relation to apron capacity. The spreadsheet model 
is again based on Blumstein’s approach. 
4.2 Apron capacity modeling 
Opposite to the case of runway system capacity, not much has been done in the field of 
macroscopic apron modeling either because its capacity was not considered a serious 
capacity constraint or the problem was too specific to be generalized. Nevertheless, a 
few generic apron models can be found in literature. They estimate dynamic apron 
capacity based on apron layout (i.e. number of stands), use strategy (with respect to 
aircraft class or airline) and weighted average stand occupancy time of the aircraft mix 
demanding service.  
These models are based on the similar approach applied to runway system capacity 
estimation, estimating the capacity of a service unit as the reciprocal of the weighted 
average occupancy time of all users served by the given service unit, as set by 
Blumstein (1959).  
In the remainder of the text, the existing analytical apron capacity estimation models are 
described, as in Mirkovic (2011b) and Mirkovic and Tosic (2013).  
In general, two different models can be found in the literature. The basic one assumes 
that all aircraft can use all available apron stands. The other assumes restriction on stand 
use by aircraft class – that aircraft of a certain class can use the stands that are designed 
for those or any larger aircraft class. 
Runway system capacity models estimate saturation capacity, which stands for the 
average number of movements that can be performed on the runway system in 1h in the 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
47 
 
presence of continuous demand, while adhering to all the separation requirements 
imposed by the ATM rules (De Neufville and Odoni, 2003). Similarly, saturation 
capacity of the apron (hereinafter referred to as “apron capacity”) can be defined as the 
average number of aircraft that can be served at the apron (fixed number of stands) 
during given period, in the presence of continuous demand (characterized by fleet mix 
and user mix), while adhering to all constraints on stand use. 
When there are no restrictions on the stand use, implying that all aircraft can use all the 
stands, the capacity of the apron can be expressed as:  
t
NC all=                  (1) 
 Nall – total number of available stands 
t – weighted average stand occupancy time of all aircraft demanding service 
∑ ⋅=
i
ii Tpt              (2) 
ip  – proportion of aircraft of class i in the population of aircraft demanding 
service 
iT  – average stand occupancy time of the aircraft of class i 
The apron capacity model with no restrictions on stand use (and numerical examples) 
can be found in relevant literature on airport planning: Horonjeff (1975), Horonjeff and 
McKelvey (1994), Ashford and Wright (1992), De Neufville and Odoni (2003), 
Horonjeff et al. (2010), Ashford et al. (2011).  
A second model assumes restrictions in stand use, by stand size. It is defined by 
Horonjeff (1975) and reformulated in later editions (Horonjeff, McKelvey, 1994; 
Horonjeff, et al., 2010). For restricted stand use, it is necessary to define the group of 
stands that can accommodate each aircraft class (classification is based on aircraft size). 
It is assumed that a stand can accommodate the aircraft class they are designed for and 
all smaller-class aircraft. The apron capacity limited by each group of stands is 
calculated from the number of stands in the group and weighted average stand 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
48 
 
occupancy time of aircraft using that group of stands. The minimum of the capacities set 
by each group of stands is the considered as the apron capacity: 
)(min ii CC =                           (3) 
The capacity limited by the group of stands available for aircraft class i (Ci) is as 
follows:  
'
'
i
i
i t
NC =               (4) 
'iN - number of stands that may be used by aircraft of class i (stands designed 
for aircraft class i and for aircraft larger than i): 
niii NNNN +++= + ...' 1            (5) 
'it - weighted average stand occupancy time demanded by all aircraft that can 
use stands from the ith group: 
∑
≥
⋅=
ik
kki Tpt '              (6) 
kp  – proportion of aircraft of class k in the population of aircraft demanding 
service 
kT  – average stand occupancy time of the aircraft of class k 
Ashford and Wright (1992) explain the apron capacity model for restricted use (by 
aircraft class), but under a different assumption. They assume that each stand can be 
used only for the aircraft they are designed for (small for small, medium for medium, 
large for large). Obviously, they do not take into account that each stand can also 
accommodate smaller aircraft classes. So, apron capacity, in their so-called “exclusive 
use” apron capacity model, is formulated as follows: 
)(min ii CC =                    (7) 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
49 
 
Where apron capacity limited by the group of stands designed for aircraft of class i is:  
ii
i
i Tp
NC ⋅=               (8) 
iN - number of stands designed for aircraft class i (only) 
ip  – proportion of aircraft of class i in the population of aircraft demanding 
service 
iT  – average stand occupancy time of the aircraft of class i 
This apron capacity model for exclusive use is not suitable for apron capacity estimation 
with restrictions on stand size/aircraft class, since the main assumption does not depict 
what is actually happening in reality. But, such an approach can be applied for capacity 
estimation of aprons with exclusive use of stands by different airlines (typical for U.S. 
airports), or by different users based on other criteria such as required security levels 
(domestic/international).  
When apron consist of the separate areas exclusively used by one user each, one can 
incorrectly conclude that the total capacity of the apron is the sum of the capacities of 
these separate areas: 
∑=
i i
i
T
NC               (9) 
However, that is not the case. To what extent each apron area is utilized depends on the 
share of users in demand ( ip ). The most restricting apron area restricts total apron 
capacity: )min( iCC = . At the same time, other areas remain underutilized.  
For a quick estimation of apron capacity U.S. FAA developed a graphical method 
shown in Figure 19 (FAA, 1983). The apron-gate capacity expressed in movements/h is 
determined as: NSG ⋅⋅* , where G* is the hourly gate capacity base, determined from 
the chart based on the share of non-widebody aircraft gate occupancy and their share in 
the fleet; S is the gate size factor, and is determined from the gate mix and percentage of 
stands that accommodate widebody aircraft; and N is the number of gates.   
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
50 
 
The most recent research, an update of the Advisory Circular (AC) 150/5060 (FAA, 
1983), did not suggest any further improvements in regard to the method or models for 
assessment of the apron (gate) capacity (TRB, 2012b).  
 
Figure 19.  FAA’s graphical method for calculating gate capacity 
In the area of analytical apron modeling there are several important papers addressing 
the issue of aircraft gate requirements estimation, which is an issue analogue to apron 
capacity estimation. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
51 
 
Bandara and Wirasinghe (1988) suggested a stochastic model to estimate the number of 
gates required to provide a given reliability, based on aircraft arrival rates at gates, the 
gate occupancy times and the aircraft separation (buffer) times, considering them as 
random variables. Mean and variance of the number of gates required was obtained 
using the moment generating function. The design number of gate was chosen to satisfy 
given reliability. Reliability is defined as the probability that there are sufficient gates to 
ensure zero delay of aircraft on the apron, during a given time period. The method is 
applicable both under common and preferential gate use strategies, as well as for 
estimating the required number of remote stands for overflow situations.  
Steuart (1974) analyzed an influence of the flights schedule and flights’ behavior 
relative to the schedule on gate requirements. The emphasis is on bank operations – “the 
cyclic exchange of a large number of scheduled gate occupancies followed by a small 
number”. Using a stochastic model based on empirical data, he showed that banking 
tends to increase the number of gates needed, observing the scheduling alternatives 
where decrease (increase) in number of aircraft in the bank is followed by decrease 
(increase) in the time interval between banks, under the constant arrival rate. He 
assumed that all flights in the bank have the same behavior. In an extension of this 
work, Hassounah and Steuart (1993) established a relationship between occupancy of 
aircraft gates and flights’ behaviors, assuming that each flight has a unique behavior 
relative to its schedule and buffer time. A stochastic model is developed to estimate the 
demand for gates as a function of time of day, under different assignment strategies 
(exclusive vs. common) and scheduling practices. They shortly discussed the influence 
of the time interval between banks on gate requirements, aiming to show how close 
banks could be scheduled without causing an excessive increase in gate requirements. 
They observed two banks of the same size, and fixed number of aircraft in the banks.  
Wirasinghe and Bandara (1990) suggested a method for determining the optimum 
number of gate positions to minimize the sum of cost of gates and cost of delays to 
aircraft. Gate cost is calculated from the number of gates and the marginal capital, 
maintenance and operating costs of the gate per day that is assumed to be constant. 
Delay cost is calculated from the average cost of delay per aircraft per hour and the total 
deterministic delay to aircraft per day caused by the lack of gates. The total 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
52 
 
deterministic delay is calculated from the maximum arrival rate, the average arrival rate 
and the time during which the average rate is exceeded. A distinction is made between 
triangular and parabolic shaped peaks when calculating delay. Cases with one and 
several non-overlapping peaks during the day were discussed, as well as the 
modification that uses a known shape of the expected arrival rate curve instead of 
average values.  
This approach was further expanded by de Barros and Wirasinghe (2004) to take into 
account: the use of common areas that could be shared by different aircraft types, and 
the effect of interest and demand increase over the lifespan of the terminal to support 
development in stages (“how much to build and when”). These issues were addressed in 
light of the introduction of new large aircraft (NLA). Two variations of the space-
sharing model were discussed. In the first one, gates for NLA and wide-body (WB) 
aircraft are allowed to be used by conventional jets (CJ). The second one assumes that 
CJ are allowed to use NLA/WB gates, and one CJ is allowed between every two WB 
and/or NLA gates. It was shown that significant savings are achievable with stage 
construction and carefully planned shared space between different aircraft types.  
Flexible design, i.e. use of shared space at airports, is addressed more often from the 
perspective of the landside. For example, De Neufville and Belin (2002) addressed use 
of shared space as a strategy to cope with traffic peaking (hourly and daily variations) 
and uncertainty (daily and long term variations). They focus primarily on shared gate 
hold-rooms to accept hourly peaking and long-term uncertainty; and shared gates to 
accept daily uncertainty (stochastic delays). Also, De Neufville (2008) addresses an 
issue of flexible design of the terminal building to cope with long-term uncertainty 
related to airport client changes (particularly the entry of low cost carriers at “legacy 
carriers airports”). 
Table 5 summarizes the most relevant literature on apron modeling with respect to: 
‐ Input required, 
‐ Output delivered, 
‐ Constraints considered, and 
‐ Impact of wave (bank) system.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
53 
 
  
Ta
bl
e 
5.
 S
um
m
ar
y 
of
 th
e 
re
le
va
nt
 li
te
ra
tu
re
 o
n 
ap
ro
n 
m
od
el
in
g 
 
pa
pe
rs
/b
oo
ks
in
pu
t
ou
tp
ut
co
ns
tr
ai
nt
s 
on
 a
/c
 ty
pe
co
ns
tr
ai
nt
s 
on
 a
ir
lin
e/
or
ig
in
‐d
es
ti
na
ti
on
w
av
e(
ba
nk
)‐s
ys
te
m
St
eu
ar
t (
19
74
)
tr
af
fic
 sc
he
du
le
ex
pe
ct
ed
 v
al
ue
 a
nd
 v
ar
ia
nc
e 
of
 
nu
m
be
r o
f a
ir
cr
fa
t o
cc
up
yi
ng
 th
e 
ga
te
s
no
t c
on
si
de
re
d
po
ss
ib
le
 a
pp
lic
at
io
n 
di
ss
cu
ss
ed
 fo
r 
ex
cl
us
iv
e 
us
e 
st
ra
te
gy
, n
ot
 
co
nn
ec
te
d 
to
 b
an
k‐
sy
st
em
co
ns
id
er
ed
 in
 d
et
ai
l ‐
 e
ff
ec
ts
 o
f b
an
k 
le
ng
th
 a
nd
 b
an
k 
se
pa
ra
tio
n,
 n
ot
 
co
nn
ec
te
d 
to
 ru
nw
ay
 p
er
fo
rm
an
ce
Ba
nd
ar
a 
an
d 
W
ir
as
in
gh
e 
(1
98
8)
av
er
ag
e 
ar
ri
va
l r
at
e,
 a
ve
ra
ge
 g
at
e 
oc
cu
pa
nc
y 
tim
e,
 a
ve
ra
ge
 
se
pa
ra
tio
n 
tim
e
re
aq
ui
re
d 
nu
m
be
r o
f g
at
es
 to
 a
ss
ur
e 
ze
ro
 d
el
ay
 w
ith
 c
er
ta
in
 re
lia
bi
lit
y
no
t c
on
si
de
re
d
pr
ef
er
en
tia
l u
se
 st
ra
te
gy
 b
y 
ai
rl
in
es
 
an
d 
ty
pe
 o
f s
er
vi
ce
s 
co
ns
id
er
ed
 a
nd
 
ap
pl
ie
d
no
t c
on
si
de
re
d
H
as
so
un
ah
 a
nd
 
St
eu
ar
t (
19
93
)
tr
af
fic
 sc
he
du
le
ex
pe
ct
ed
 v
al
ue
 a
nd
 v
ar
ia
nc
e 
of
 
nu
m
be
r o
f a
ir
cr
fa
t o
cc
up
yi
ng
 th
e 
ga
te
s
co
ns
id
er
ed
 a
nd
 a
pp
lie
d 
w
ith
ou
t 
co
ns
tr
ai
nt
s o
n 
ai
rli
ne
/t
yp
e 
of
 s
er
vi
ce
 
co
ns
tr
ai
nt
s
ex
cl
us
iv
e 
us
e 
st
ra
te
gy
 w
ith
ou
t 
co
ns
tr
ai
nt
s o
n 
a/
c 
ty
pe
 c
on
si
de
re
d 
an
d 
ap
pl
ie
d
po
ss
ib
le
 a
pp
lic
at
io
n 
di
sc
us
se
d
W
ir
as
in
gh
e 
an
d 
Ba
nd
ar
a 
(1
99
0)
av
er
ag
e 
ar
ri
va
l r
at
e,
 m
ax
 a
rr
iv
al
 
ra
te
, a
rr
iv
al
 ra
te
 c
ur
ve
s,
 a
ve
ra
ge
 
ga
te
 o
cc
up
an
cy
 ti
m
e,
 a
ve
ra
ge
 
se
pa
ra
tio
n 
tim
e
re
aq
ui
re
d 
nu
m
be
r o
f g
at
es
  b
as
ed
 o
n 
th
e 
op
tim
um
 a
rr
iv
al
 ra
te
 th
at
 
m
in
im
iz
es
 d
el
ay
+g
at
e 
co
st
no
t c
on
si
de
re
d
po
ss
ib
le
 a
pp
lic
at
io
n 
di
sc
us
se
d 
fo
r 
pr
ef
er
en
tia
l u
se
 st
ra
te
gy
no
t c
on
si
de
re
d
de
 B
ar
ro
s a
nd
 
W
ir
as
in
gh
e 
(2
00
4)
se
rv
ic
e 
ra
te
, a
ve
ra
ge
 a
rr
iv
al
 ra
te
, 
m
ax
 a
rr
iv
al
 ra
te
, a
ve
ra
ge
 g
at
e 
oc
cu
pa
nc
y 
tim
e,
 a
ve
ra
ge
 
se
pa
ra
tio
n 
tim
e
re
qu
ir
ed
 n
um
be
r o
f g
at
es
 (t
er
m
in
al
 
fr
on
ta
ge
 le
ng
th
) b
as
ed
 o
n 
th
e 
op
tim
um
 a
rr
iv
al
 ra
te
 th
at
 m
in
im
iz
es
 
de
la
y+
ga
te
 c
os
t
co
ns
id
er
ed
 in
 d
et
ai
l (
di
ff
er
en
t s
pa
ce
 
us
ag
e 
st
ra
te
gi
es
 a
dd
re
ss
ed
) w
ith
ou
t 
an
y 
ot
he
r c
on
st
ra
in
ts
 (o
n 
ty
pe
 o
f 
tr
af
fic
/a
ir
lin
e)
no
t c
on
si
de
re
d
no
t c
on
si
de
re
d
H
or
on
je
ff
  e
t a
l. 
(1
97
5‐
20
10
)
nu
m
be
r o
f g
at
es
, e
xp
ec
te
d 
(w
ei
gh
te
d)
 g
at
e 
oc
cu
pa
nc
y 
tim
e 
ga
te
 c
ap
ac
ity
 
no
t c
on
si
de
re
d
ex
cl
us
iv
e 
us
e 
st
ra
te
gy
no
t c
on
si
de
re
d
A
sh
fo
rd
 a
nd
 
W
ri
gh
t (
19
92
)
nu
m
be
r o
f g
at
es
, (
w
ei
gh
te
d)
 
av
er
ag
e 
ga
te
 o
cc
up
an
cy
 ti
m
e 
ga
te
 c
ap
ac
ity
co
ns
id
er
ed
 (g
at
es
 a
re
 a
va
ila
bl
e 
fo
r 
de
si
gn
 a
ir
cr
fa
t a
nd
 a
ny
 s
m
al
le
r 
ai
rc
rf
at
) w
ith
ou
t o
th
er
 c
on
st
ra
in
ts
 (o
n 
ty
pe
 o
f t
ra
ff
ic
/a
ir
lin
e)
no
t c
on
si
de
re
d
no
t c
on
si
de
re
d
FA
A
 (1
98
3)
 
gr
ap
hi
ca
l 
ap
pr
oa
ch
nu
m
be
r o
f g
at
es
, g
at
e 
si
ze
 fa
ct
or
 
an
d 
ho
ur
ly
 g
at
e 
ca
pa
ci
ty
 b
as
e
ho
ur
ly
 c
ap
ac
ity
 o
f g
at
es
 (g
ra
ph
ic
al
 
ap
pr
oa
ch
) ‐
 q
ue
st
in
ab
le
 re
su
lts
co
ns
id
er
ed
no
t c
on
si
de
re
d
no
t c
on
si
de
re
d
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
54 
 
4.3 A way forward 
The thesis offers modifications/extensions to apron capacity modeling and, together 
with that, an overall airside capacity analysis. It considers the following: 
- Both aircraft class and user constraints and their impact on apron capacity. 
- The impact of wave-system at hub airports on apron capacity. It does not derive 
it from the detailed traffic schedules, but observes apron capacity sensitivity to 
different hubbing parameters.  
- The functional relationship between the runway system and the apron(s), in 
order to provide an overall airside capacity picture by connecting existing 
runway system models to modified apron models.  
Furthermore, it observes reserve apron capacity, giving an opportunity to gain 
additional capacity with available resources reallocation prior to physical expansion. 
Reserve capacity is observed through the concept of apron capacity flexibility from the 
airside perspective, accounting primarily for operational users’ constraints. 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
55 
 
5 An approach to analyzing overall airport airside capacity 
As a reflection of what occurs in the case of major airports worldwide today, the runway 
system is considered to be the main capacity constraint. This is why airside capacity is 
usually expressed through runway system capacity. At the same time, complex taxiway 
systems and huge apron(s), with a large number of remote stands, usually operate with 
spared capacity. The literature overview on airside modeling (Chapter 4) confirms that 
the most mature area is the runway system capacity assessment, together with 
simulation modeling of the entire airside. The issue of the apron capacity assessment is 
not often referred to in the literature.   
However, the air transport network counts dozens of major airports while, at the same 
time, there are hundreds or even thousands of small-to-medium airports that suffer from 
different capacity issues. At these airports, which mainly (but not necessarily) operate 
with single or two dependant runways, the capacity constraint is often at the terminal 
complex, until runway system capacity limit is reached27. In order to accept traffic 
increase, the airport needs to rearrange/expand terminal-apron complex, maintaining 
balanced usage of the runway system, all until the capacity issue switches back to 
runway system. After (if) another runway is built, again the same task lays before the 
airport planners, to further develop other elements in accordance with traffic changes.  
An approach to analyzing the overall airside capacity of the existing (built) system 
under given physical and operational constraints and for given demand characteristics, 
is proposed in Section 5.1. Runway system capacity is assumed as known i.e. possible 
to obtain through any of existing runway system capacity models. Taxiway system is 
assumed not to be the capacity issue. The main attention is on the apron capacity 
modeling. Existing apron capacity estimation models are modified/expanded, as given 
in Sections 5.2. In Section 5.3 sensitivity of the apron capacity on demand structure is 
represented using apron capacity envelopes. Section 5.4 elaborates important variables 
that influence apron capacity estimates.  
                                                 
27 A single runway at Gatwick Airport handles between 240,000 and 260,000 annual movements since 
2000 (in 2007 it handled 266,550 movements). 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
56 
 
As summarized in Chapter 3, although runway system and apron capacity are affected 
by similar factors, it does not imply that both of those resources reach their maximum 
throughput under the same conditions, because the nature of processes is different, due 
to the different roles of the runway system and the apron in the airside system. In order 
to create an overall airside capacity usage scheme, it is necessary to understand and 
define functional relationship between the runway system and the aprons.  
The presence of different market segments at the airport can influence this functional 
relationship, due to different usage of the resources. As it is discussed in Chapter 1, the 
runway system is usually the common resource for all, which is not the case with 
apron(s), where different market segment may have their own terminal-apron 
complexes or preferential apron areas. In such case, the airport represents a system at 
which all clients enter/exist through the same service unit, and are processed at different 
service units.  
The thesis deals only with airports with a dominant passenger market segment. As 
summarized in Chapter 1, the majority of European airports have only one dominant 
market segment, of which 85% are passenger dominated. Among all European airports, 
regardless of the number of dominant market segments, again the majority (72%) serve 
mainly passenger traffic (Eurocontrol, 2007). 
In the case of passenger market segments (traditional, low-cost and/or charter) all three 
may share the same terminal building, while certain segmentation usually exist on the 
apron, for example preference of the LCCs and charter airlines to use remote over the 
contact stands. Additional segmentation is not expected to have any impact on the 
functional relationship between apron(s) and the runway system, because all aircraft are 
of the equal priority when using the runway system, unlike the case when different 
market segments are involved.  
At passenger airports runway-apron functional relationship is mainly influenced by the 
role of the airport in the air transport network (elaborated in Chapter 2). Section 5.5 
addresses the difference in capacity utilization of available resources, depending on the 
nature of the traffic at the airport. Two main airport types, with respect to their role in 
air transport network, are analyzed: origin-destination airports, serving primarily point-
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
57 
 
to-point traffic (resulting in rather “uniform” traffic distribution during the day, with 
more or less pronounced peak periods) and hub airports serving primarily transfer traffic 
typically concentrated in waves (banks) of flights. Following network evolution the 
thesis also addresses the co-existence of coordinated and point-to-point flights at a 
single airport. Different examples are used to show how apron and overall capacity react 
to changes in: type of traffic served, runway system performance, and apron operational 
constraints.  
The thesis also addresses the issue of reserve/latent apron capacity (Section 5.6). 
Namely, capacity bottleneck at the apron is not necessarily a consequence of capacity 
shortage, but it may appear due to a mismatch between demand structure and constraints 
on available resources allocation. Reserve capacity is discussed through the concept of 
apron capacity flexibility. It is suggested how to express and interpret apron capacity 
flexibility. On one hand, apron capacity flexibility is observed as an indicator of ability 
of the apron to respond on changes in demand structure. At the same time, it is an 
indicator of latent/reserve capacity that may be “activated” by reallocation of available 
resources i.e. relaxation of the constraints on stand use. 
Having the capacity of both elements and their functional relationship, under given 
demand characteristics, it is possible to create an overall capacity usage scheme, to 
identify bottlenecks and underutilized resources within the system. The discussion is 
included through different examples. 
At the end (Section 5.7) a short discussion is given on possible areas of application of 
the proposed approach in the field of airside capacity planning.  
5.1 Conceptual model 
An approach to analyzing airport airside capacity and how it is utilized with respect to 
demand characteristic (structure and distribution in time) is given in the flowchart in 
Figure 20. 
The thesis addresses mainly issues related to apron capacity estimation and the impact 
of different traffic patterns of capacity utilization which are identified as the main gaps 
in current literature.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
58 
 
However, runway system capacity is needed in order to further discus its connection 
with apron capacity, but it is assumed in this thesis as known.  
An analysis should be based on representative peak periods reflecting the traffic 
structure of the typical day(s) at the airport. According to peak period discussion in 
Chapter 2, it is suggested to observe typical 2.5-3.5h period(s) for O/D airports, while 
for hubs wave length and wave repeat cycle are far more important.  
 
Figure 20. An approach to overall airside capacity analysis 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
Describe system design and 
operating conditions  
(layout and constraints on 
runway and apron usage; 
occupancy times, buffers, etc.) 
Describe RWY-APR 
functional relationship 
in respect to demand 
characteristics 
Describe demand 
structure  
(fleet mix, user 
mix, etc.) 
Describe demand 
characteristics with 
respect to nature of 
traffic at the airport 
(hub, origin-destination) 
Estimate APR 
capacity using newly 
developed model 
Estimate RWY 
capacity using any  
of existing models  
“Merging” through RWY-APR 
functional relationship  Identification  of bottlenecks/ 
underutilized 
resources 
Analyze potential to respond to traffic 
structure changes through different 
operational schemes (e.g. reallocation of 
parking stands between different users to 
gain additional apron throughput)  
Create overall capacity usage scheme 
under given demand. Identify conditions 
under which bottlenecks “move“ from 
one resource to another.
Reserve 
capacity 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
59 
 
The models for assessing and analyzing apron capacity and overall airside capacity 
proposed in the thesis are based on the following assumptions: 
‐ The proposed analytical models, representing extension and modification of the 
existing models, estimate apron saturation capacity. This implies continuous 
demand for service during the specified period of time. 
‐ The apron capacity models are deterministic, implying that the average values of 
particular variables are used in estimating capacity. 
‐ The sensitivity of apron capacity to demand structure changes is observed from 
the perspective of the dominant user (vs. all other users). 
‐ The runway system capacity is known, i.e. estimated through any of the existing 
runway capacity models. The taxiway system is not considered the potential 
bottleneck in the airport airside area, and it is not considered when defining the 
runway-apron functional relationship.  
‐ The focus is on the two types of passenger airports: O/D airports serving mainly 
point-to-point traffic and hub airports, serving coordinated flights of the base 
airline, primarily or in combination with additional point-to-point traffic. The 
presence of other market segments at the airport is not considered. 
‐ Parameters related to and derived from wave-system structure are based on the 
ideal wave: arrival and departure time windows are of approximately the same 
length and the sequence of arrivals is the same as the sequence of departures.  
‐ Analysis of the potential for providing additional apron capacity through 
relaxation of the user constraints does not take into consideration associated 
costs for implementation of the proposed solutions.  
5.2 Apron capacity estimation module 
Based on the existing models (as given in Section 4.2), an extension is proposed, that 
combines size and user restrictions, in order to include both physical and operational 
constraints on aircraft stand usage.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
60 
 
Apron capacity is determined by the minimum of the capacities set by each ijth group of 
stands, (Mirkovic, 2011b; Mirkovic and Tosic, 2013): 
( ) ⎟⎟⎠
⎞
⎜⎜⎝
⎛==
'
'
minmin
ij
ij
ijijij t
N
CC           (10) 
where: 
 i designates the user, [ ]ni ,1∈  
j designates the aircraft class, [ ]mj ,1∈  where 1 is the smallest aircraft class, and 
m is the largest aircraft class 
'ijN - number of stands that may be used by aircraft of user i and class j (stands 
allowed to be used by user i, designed for aircraft class j and for aircraft larger 
than j)  
∑∑
∈ ∈
=
Kk Ll
klij NN '                        (11) 
{kK = [ ]nk ,1∈  and user-class k allows its stands to be used by user-class 
}i , [ ]nK ,1⊆  
{lL = [ ]mj ,1∈  and aircraft class l is equal or larger than aircraft class j, 
}jl ≥ , i.e. [ ]mjl ,∈  
 
'jit  - weighted average stand occupancy time demanded by all user/aircraft 
class combination allowed to use the ijth group of stands 
∑∑
∈ ∈
⋅=
Kk Ll
klklij Tpt '                                (12) 
The same sets of users (K) and aircraft classes (L) apply as in expression 
(11), for 'ijN  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
61 
 
klp  – proportion of aircraft of user k and class l in the population of aircraft 
demanding service 
klT  – average stand occupancy time of the aircraft of user k and class l  
ijC - apron capacity limited by the group of stands available for user i and 
aircraft class j 
For aircraft classes standard ICAO (International Civil Aviation Organization) 
categorization may apply (from A to F), or they may be categorized in a different way 
e.g. small, medium and large. 
A user-class criterion depends on the airport itself. Considering only passenger traffic, it 
may be airlines/alliances (typical U.S. stand usage strategy - exclusive, preferential, 
joint), and/or type of service with respect to security level required inside the terminal 
building (general case - domestic/international, or European specific – domestic, 
Schengen, non-Schengen, international with special requirements).  
For example, the airport is operated by base airline, other traditional airlines and low-
cost airlines, on domestic and international routes. It results in the following users: base-
domestic, base-international, base-mix, other-domestic, other-international, other-mix, 
LCC-domestic, LCC-international and LCC-mix. Not all of them necessarily account 
for a significant share in peak period. For example, assuming that other-domestic, other-
mix and LCC-mix account for very small shares, it leaves six (out of nine) 
representative users. 
The number of stands that may be used by user class i and aircraft class j is determined 
from apron layout and operational constraints defined by: 
‐ Total number of aircraft stands; number of contact stands (i.e. gates) and remote 
stands; 
‐ Number of stands by aircraft class, assuming that the stand may be used by 
design aircraft and any smaller aircraft;  
‐ Number of stands by user - it applies particularly to contact stands, although 
some preferences may exist in practice towards remote stands as well; and  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
62 
 
‐ Policy of aircraft stands use - set of rules that define if the group of stands of a 
certain user may or may not to be used by other users. 
Weighted average stand occupancy time is calculated from average stand occupancy 
times for different user/aircraft class combination and demand structure. Demand 
structure considers: 
‐ Share of different aircraft classes, and  
‐ Share of different users.  
Numerical example from AC 150/5060 (FAA, 1983) is used to compare results from the 
analytical model to results from the FAA’s graphical method (Mirkovic, 2011b). The 
results are given in Appendix 10.  
For apron(s) with flexible layouts, with respect to aircraft classes, combined with 
different user constraints (either by airline, or type of service) the calculation can be 
somewhat tedious. It requires several iterations, as it is explained in Appendix 11. 
5.3 Apron capacity representation 
The common way to illustrate saturation capacity for a given configuration of the 
runway system and given demand structure is the runway capacity envelope, introduced 
by Gilbo (1993). He uses the runway capacity envelope to represent a set of capacity 
values that reflect the operational capability of the airport under certain conditions. An 
example is given in Figure 21, for a single runway.  
Runway capacity envelope consists of four typical points, representing different 
arrival/departure shares. Point 1 represents the capacity of arrivals-only. Point 2 
represents so-called departure-free capacity and considers additional departures that can 
be performed without any changes in arrivals separation. Point 3 represents capacity 
under a 50/50% arrivals/departures share, and Point 4 is the departures-only capacity. 
The thesis suggests possible shape(s) of the apron capacity envelope to illustrate the 
ability of a certain apron configuration to accept different demand structures with 
respect to dominant users, rather than using a single number instead (Mirkovic, 2011b; 
Mirkovic and Tosic, 2013).  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
63 
 
 
 
Figure 21. Example of the runway capacity envelope for a single runway 
A runway capacity envelope gives the saturation capacity for one runway system 
configuration, for a given demand structure, but for different shares of arrivals and 
departures. Runway system capacity is expressed in operations per hour, but smaller 
time units are also used (usually 15min). If we consider that arrivals and departures are 
the main “customers” of the runway system as a resource, conversely, the apron 
capacity envelope should represent the saturation capacity of a certain apron 
configuration and given demand structure (fleet mix, share of different aircraft classes) 
for different shares of its users (airlines, origin/destination combination, traffic type, 
etc.). Apron capacity is expressed in aircraft per hour. Depending on the traffic pattern, 
different time units might be more suitable than one hour, e.g. the length of the arrival 
wave at hub airports, as discussed earlier. In Figure 22 the apron capacity envelope for 
Example 1 is given.  
0
5
10
15
20
25
30
35
40
45
50
55
60
0 5 10 15 20 25 30 35 40 45 50 55 60
de
pa
rtu
re
s/
h
arrivals/h
single runway 
1
3
4
2
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
64 
 
Example 1: An apron has 11 stands, of which 5 stands are available only for domestic 
rotations28 (both origin of flight and destination after turnaround are domestic) and 6 
stands are available for all rotations (domestic, international and mixed). Average stand 
occupancy time for domestic rotations is 45min, and for other than domestic rotations 
(domestic-international, international-international) it is 55min.  
If we assume demand of 50% domestic flights and 50% other than domestic flights, 
apron capacity is 13.1 aircraft/h29. Apron capacity changes with user structure changes 
in demand, as shown in Figure 22.  
 
Figure 22. Apron capacity envelope in Example 1 
Together with apron configuration changes, the shape of the apron capacity envelope 
also changes. Figures 23 and 24 represent a set of apron capacity envelopes for different 
apron configurations. Number of stands available for domestic flights and number of 
stands for all flights is varied (the total number of stands remains the same).  
                                                 
28 A rotation refers to a combination of the aircraft origin and its destination after its turn-around. 
29 ( ) ( ) haircraftCCC /1.131.13,2.13min60
555.0
6,60
555.0455.0
65min,min 21 ==⎟⎠
⎞⎜⎝
⎛ ⋅⋅⋅⋅+⋅
+==  
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
0-100 10-90 20-80 30-70 40-60 50-50 60-40 70-30 80-20 90-10 100-0
ap
ro
n 
ca
pa
ci
ty
 (a
irc
ra
ft
/h
)
domestic/other f lights share (%)
5 dom.f l. stands + 6 ALL f l. stands
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
65 
 
 
Figure 23. Set of apron capacity envelopes illustrating an increase in number of mixed-use stands in 
Example 1 
 
Figure 24. Set of apron capacity envelopes illustrating an increase in number of exclusive-use stands in 
Example 1 
0
2
4
6
8
10
12
14
16
0-100 10-90 20-80 30-70 40-60 50-50 60-40 70-30 80-20 90-10 100-0
ap
ro
n 
ca
pa
ci
ty
 (a
irc
ra
ft
/h
)
domestic/other f lights share (%)
5Dom/6All 4Dom/7All 3Dom/8All
2Dom/9All 1Dom/10All
0
2
4
6
8
10
12
14
16
0-100 10-90 20-80 30-70 40-60 50-50 60-40 70-30 80-20 90-10 100-0
ap
ro
n 
ca
pa
ci
ty
 (a
irc
ra
ft
/h
)
domestic/other f lights share (%)
5Dom/6All 6Dom/5All 7Dom/4All
8Dom/3All 9Dom/2All 10Dom/1All
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
66 
 
Figure 23 shows that, as the number of stands for all flights increases, the apron 
becomes less sensitive to user mix in total demand. In extreme case of a configuration 
with one stand for domestic flights and 10 stands for all flights, the capacity ranges from 
10.9 to 14.7 aircraft/h. As the number of “exclusive” stands for domestic flights 
increases, the apron becomes more sensitive to the share of different users in total 
demand, as it is shown in Figure 24. In the opposite extreme case, assuming a 
configuration with 10 stands for domestic flights and only one stand for all flights, the 
capacity of the apron ranges between (only) 1.1 aircraft/h and 14.7 aircraft/h. 
This is a simple example to show an apron’s ability to accept different demand 
structures, and how it changes with apron configuration changes. In this example, 
demand structure is given as the share of domestic/other flights. Depending on the 
policy of stand usage at the apron it can be expressed with respect to other users, such as 
airlines/alliances. 
The indicator of the sensitivity of a certain apron configuration to demand structure 
changes with respect to users will be referred to, hereinafter, as apron capacity 
flexibility. With an increase in the number of exclusive use stands, apron capacity 
flexibility decreases. Higher flexibility goes with apron configurations with fewer 
constraints on stand usage.  
“The knowledge of the sensitivity of a physical facility component to a variation in 
demand can lead to better decisions and an understanding of the flexibility in any 
facility design” (Horonjeff et al., 2010). 
Saleh et al. (2009) thoroughly discuss the concept of flexibility in different disciplines 
(among others engineering design). They argue that “both robustness and flexibility of a 
design refer to the ability of a system to handle change”, but “what is changing” and 
“how is change achieved” differs. They discuss views of different authors on robustness 
and flexibility in design, and accept the following as the most appropriate: 
‐ Robustness of a design “is the property of a system that allows it to satisfy a 
fixed set of requirements, despite changes in the environment or within the 
system”. It “implies de-sensitizing the system’s performance or quality 
characteristics to changes in the system environment, or within the system”.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
67 
 
‐ Flexibility of a design “implies an ability of the design to be changed in order to 
track requirement changes”. 
Authors consider flexibility and robustness to be separate characteristics of a design.  
Robustness of the apron may be used to describe its “ability to handle change” when 
speaking of e.g. apron design with respect to minimum safe separation (taxiway to 
object, aircraft on stand to other aircraft and/or object, etc.), or apron surface strength. 
In order to describe the ability of the apron to accept changes in demand structure in 
terms of capacity, the term flexibility is more appropriate. In this case, “apron” design is 
defined by set of operational rules, which allow greater or fewer possibilities to change 
in order to respond to demand changes. 
Flexible use of the same area by different aircraft sizes was addressed by De Barros and 
Wirasinghe (2004). They propose an analytical methodology to determine the number 
of contact stands (gates) required to accommodate a mix of aircraft classes, which takes 
into account the concept of space sharing between different aircraft classes. They 
determine shared space based on the requirements for gates by each aircraft class in 
peak periods and idle gate positions during secondary peaks of a typical day. The space 
to be shared should be carefully planned so that the apron remains able to respond 
successfully to variations in demand (volume and structure) during a typical day. With 
space sharing, the total area required to serve the demand is reduced, as it becomes 
more flexible with respect to class of aircraft. In addition to aircraft class, the paper also 
discusses operational issues, imposed by type of flights/passengers, which needs to be 
resolved within the terminal, so as to enable full implementation of space sharing.  
If particular operational constraints (different types of users, aircraft class) are 
combined, there are many possible scenarios that could be analyzed. However, not all of 
them are within the scope of airport planner’s interest. For the purpose of comparison 
between different scenarios, a set of selected segments of the apron capacity envelopes 
can be observed. 
Example 2: An airport has an apron of 11 stands, of which 5 stands are exclusively used 
by the base airline and 6 stands are for all other airlines. At the base airline apron area, 3 
stands are designed for aircraft class 1, and 2 stands for aircraft class 2. The apron area 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
68 
 
for other airlines consists of 1 stand for aircraft class 1, 2 stands for aircraft class 2 and 
3 stands for aircraft class 3.  
The share of flights (by airline and aircraft class) and average occupancy times are 
given in Table 6. The share of the base airline is 40% and the share of other airlines is 
60%. 70% of base airline flights are operated by aircraft class 1 and 30% by aircraft 
class 2. 15% of the flights of other airlines are operated with aircraft class 1, 50% by 
aircraft class 2 and 35% by aircraft class 3.  
Table 6. Demand structure and average stand occupancy times in Example 2 
 
Let us assume that an increase in base airline flights is predicted for the future, as well 
as a change in fleet mix. Three expected demand scenarios are: 50/50, 60/40 and 70/30 
shares of base/other airlines in traffic. Fleet mix by user is given in Table 7. 
Table 7. Current and future demand structure with respect to fleet mix and users in Example 2 
 
Three possible Scenarios are considered as a response to the future change in demand. 
Scenario 1 assumes changes in layout at the base airline apron area. One small stand for 
aircraft class 1 is widened to accept aircraft class 2, resulting in 2 small and 3 mid-sized 
stands. Scenario 2 assumes changes in operational constraints. The layout is the same as 
in the current state, but all stands in apron area for other airlines are available for the 
base airline. Scenario 3 includes both changes in layout (the same as in Scenario 1) and 
in operational constraints (as in Scenario 2).  
number of 
stands user
aircraft 
class
share in 
demand (%)
average stand 
occupancy time 
(min)
3 1 28 30
2 2 12 45
1 1 9 30
2 2 30 45
3 3 21 70
other 
airlines
base airline
user aircraft class
current demand 
structure (%)
future demand 
structure (%)
1 70 55
2 30 45
1 15 20
2 50 45
3 35 35
base airline
other airlines
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
69 
 
Apron capacity envelopes (not complete, but relevant segments) for the basic and three 
proposed scenarios are given in Figure 25.   
 
Figure 25. Relevant segments of apron capacity envelopes for scenarios analyzed in Example 2 
The solid line in Figure 25 shows how the existing apron (basic scenario) reacts to 
changes in demand. For the 50/50% share it will provide somewhat higher capacity (by 
11%) than with the current demand structure (under which apron area for base airline is 
underutilized). With a further increase of the base airline share, apron capacity falls 
below the current capacity level (dashed horizontal line in Figure 25), and continues 
decreasing with an increase in the base airline share. Changes in apron capacity (in %) 
relative to the current capacity level (10.6 aircraft/h) are given in Table 8 (Column 2 for 
the Basic Scenario). 
In all three base/other airlines share cases, Scenario 1 provides higher apron capacity 
than the current level and the Basic Scenario, for given demand structures (Column 3 in 
Table 8). Although better than the Basic Scenario, Scenario 1 responds well on an 
increase in base airline share up to certain level, after which a mismatch between the 
6
7
8
9
10
11
12
13
14
15
16
17
50/50 60/40 70/30
ap
ro
n 
ca
pa
ci
ty
 (a
irc
ra
ft
/h
)
base/other airlines share (%)
basic scenario1 scenario2
scenario3 current AC
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
70 
 
share of stands by the users and share of users in demand appears, and apron capacity 
starts to decrease.  
Table 8. Apron capacity changes (in %) relative to the current apron capacity level (10.6 aircraft/h) in 
Example 2 
 
In Scenario 2 the apron capacity is similar to Scenario 1 for the 50/50 and 60/40 mixes 
of base/other airlines. With a more significant increase in base airline flights (70/30 
mix), Scenario 2 is much better than Scenario 1 due to more flexibile use of available 
apron stands. For the case of 70/30 mix, Scenario 2 provides capacity which is for 43% 
higher than current capacity (Column 4 in Table 8), against the 10% increase which is 
gained in Scenario 1.  
As expected, the most significant improvement can be achieved in Scenario 3, which 
considers both changes in layout (the same as in Scenario 1) and operational constraints 
(as in Scenario 2). Scenario 3 provides a capacity increase from 26% to 52% (Column 5 
in Table 8) depending on the mix of aircraft/flights of the base and other airlines. 
5.4 Utilization factor and stand occupancy times 
The calculations as given above assume that all stands are fully utilized (i.e. 100% of 
time). In order to get more realistic estimate of apron capacity, calculated values can be 
corrected by introducing the stand utilization factor. It represents the amount of time the 
stands are occupied with respect to total time available (Horonjeff, 1975; Bandara and 
Wirasinghe, 1988; Hassounah and Steuart, 1993; De Neufville and Odoni (2003). 
In general, it is assumed that, for each group of stands, the following applies: the time 
supplied by the stands has to be greater or equal to the time demanded for the same 
stands. For the ith group of stands it follows (Horonjeff and McKelvey, 1994): 
iiii CtN ''' ≥µ             (13) 
base/other 
airline share Basic scenario Scenario 1 Scenario 2 Scenario 3
50/50 11 26 21 26
60/40 -7 28 31 46
70/30 -20 10 43 52
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
71 
 
In expression (13): 
'iµ  represents the percentage of time in an hour that the stands from the ith group 
can be used by all aircraft that are allowed to use stands from the ith group 
'iN - number of stands in the ith group of stands 
'it  - weighted average stand occupancy time demanded by all aircraft allowed to 
use the ith group of stands 
iC - apron capacity limited by the ith group of stands  
Consequently, apron capacity corrected for utilization factor is: )min( iCC =
(expression 3) where restriction set by the ith group of stands is: 
'
'
'
i
i
i
i t
NC µ⋅=             (14) 
Up to date experience has shown that the utilization factor typically ranges from 0.6 to 
0.8 (Ashford et al., 2011). It becomes more complicated when it has to be expressed for 
different groups of stands (by user/size), where one stand can belong to more than one 
group.   
Another approach to avoid overestimating of apron capacity takes into account that the 
time during which a stand is blocked by one aircraft and cannot be used by any other, 
consists not only of the turnaround time at the stand, but also includes the separation 
time between two consecutive aircraft using the same stand.  
Bandara and Wirasinghe (1988) define separation time as the time between a departure 
from a gate position and the next arrival. It consists of push-out or power-out time, the 
time required by departing aircraft to clear the apron, and the time required by arriving 
aircraft to move in from the apron entrance to the gate position. Defined like this, 
separation time depends on the apron and terminal layouts and it is independent of flight 
schedules. On the other hand, the utilization factor, determined empirically, is a function 
of number of stands and existing schedule at the airport where it is estimated. Because 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
72 
 
of that, Bandara and Wirasinghe (1988) considered separation time to be more 
convenient correction than utilization factor.  
De Neufville and Odoni (2003) define so-called stand blocking time as the sum of stand 
occupancy time (SOT), positioning time and buffer time. In their approach SOT stands 
for scheduled turnaround time. It depends on the size of the aircraft, flight distance, 
airline, and model (low-cost, traditional, general aviation, etc.). It can range from 20min 
(for small regional aircraft) to 4h (for wide-body intercontinental flights). According to 
authors, positioning time can range from 2min to 10min depending on whether the 
aircraft does power in /power out or it is pushed-back. In order to absorb possible 
disturbances in the flight schedule they argue it is desirable by airports (and airlines) to 
provide the buffer time between two consecutive users of the same stand. It can range 
from several minutes to an hour, depending on the local circumstances, such as: typical 
delays, policy of stand-use, stand type (contact or remote), apron configuration, etc.  
For a certain user, as is the case for low-cost airlines, one can make a good guess of 
average turnaround times. Their business model is such that it insists on short 
turnaround times (25min to 30min), operating primarily point-to-point services with 
uniform fleet on short-to-medium haul routes, without being dependant on any 
connecting flights, at least not intentionally. On the other hand, traditional airlines, 
operating both hub-and-spoke and point-to-point concept, with variety of aircraft classes 
and wide range of routes, may have turnaround times from 20min to several hours 
(TRB, 1987; de Neufville and Odoni, 2002).  
The analysis of turnaround times and important factors that affect them (e.g. 
manufacturers’ requirements30, scheduling, hubbing, increasing aircraft size, etc.) is 
elaborated by Caves (1994). However, it is based on empirical analysis using data that is 
too old to be considered relevant today. For example it involves old Munich Airport – 
Riem and effects of hubbing during the 1990s, which have significantly evolved in the 
meantime. Average turnaround times showed tendency to decrease when the hubbing 
role of an airport increases. That is exactly the opposite if we assume temporal 
                                                 
30They are usually based on assumptions, such as: simultaneous forward and aft galley servicing even for 
small aircraft, fueling simultaneously with passenger embarking and disembarking, etc., which present 
turnaround times as being shorter than it is possible to achieve under real operating conditions. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
73 
 
concentration of traffic in waves, which may increase only “ground” times, aiming to 
connect with all flights within the wave. In his lecture notes Dennis (2005), on a simple 
example, argues that the average turnaround time increases together with the increase in 
the number of aircraft per wave. 
Different information on apron capacity is provided when utilization factor and 
separation times are employed as a correction As discussed above, using separation 
times provides information on available capacity of the given apron layout regardless of 
the flight schedule. On the other hand, the utilization factor is entirely a reflection of 
demand requirements, and it provides information to what extent available capacity is 
utilized under given demand.  
When using separation time and/or the utilization factor, one should be cautious, in 
order not to include the same correction twice in the calculation which would result in 
underestimating the capacity of the apron. For example, if the utilization factor is 
expressed considering separation time and/or buffer time between two occupancies of 
the same apron area, then the same should not be added again to the stand occupancy 
time.  
In this thesis apron capacity estimation is based on the stand occupancy times that 
account for:  
‐ The turnaround time for different users/aircraft classes and separation time 
between two consecutive occupancies of the same apron area (for O/D airports);  
‐ The time required for facilitating transfers between connecting flights, and/or the 
time between consecutive waves of the same airline/alliance (at hub airports).  
Stand occupancy time reflects the time during which a stand is reserved/blocked for a 
particular user regardless whether the aircraft (physically) occupies it during entire time. 
It is desirable to select representative user/aircraft class categories carefully, in order to 
catch the variety of turnaround times typical for given airport.  
Separation time considers both the time that has to be provided between two 
consecutive occupancies of the same group of stands, due to apron layout, and, if it 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
74 
 
applies to the airport, additional buffer time for absorbing regular disturbances in traffic 
schedules. Certain buffer times have been already included by airlines in their planned 
turnaround times. 
At hub airports, the stand occupancy time for aircraft concentrated in waves of flights 
has to account for the time required to enable for the efficient transfers between flights 
and/or the time between two consecutive waves. Apron utilized capacity can be derived 
by applying both to the hub case serving only connecting flights, aiming to present to 
what extent theoretical capacity may be exploited depending on the specific demand for 
service (airline concentrate its fights in waves). In general, the period during which a 
stand is blocked by connecting flights, preventing the same resources to be used by 
other potential users, depends on the strategy of apron usage applied at the airport.  
The influence of waves of flights at hub airports on stand occupancy/blocking times and 
consequently apron capacity is addressed in details in the reminder of this chapter. 
5.5 Runway-apron functional relationship31 under different traffic patterns 
Services provided to aircraft on the runway system and in the apron(s) (i.e. aircraft 
contact and remote stands) are different in nature. The runway system is entry/exit point 
to/from the system, where service times are the order of magnitude of a few minutes. At 
apron(s) aircraft are turned around which requires service times from 20min to as much 
as several hours. Interaction of arrival and departure flows exists at both elements. 
Different arrivals and departures interact at these two service units, due to difference in 
service times and the transitional (taxi) times between them. 
Analytical models express apron capacity in aircraft/h, while the runway system 
capacity is expressed in operations/h.  
The roughest calculation is to multiply aircraft/h by two, assuming that one aircraft is 
related to two operations – arrival and departure. Such a calculation is used, for 
example, in the FAA’s graphical method.  
                                                 
31 The physical runway-apron relationship related to taxi times and their impact on exchange of arrivals 
and departures between apron and runway is not considered in this thesis.  
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
75 
 
De Neufville and Odoni (2003) take into consideration largest fraction of arrivals in the 
traffic mix during a certain time interval. For example, if we have 65% of arrivals and 
35% of departures and an apron capacity of 11 aircraft/h, that will correspond to 16.9 
operations/h.  
However, this calculation is still not sufficient to capture the real connection between 
apron and the runway system capacities, since this relation can be rather complex. It 
generally depends on factors such as: dominant market segments (e.g. scheduled, 
charter, low-cost, general aviation), type of airport with respect to traffic pattern 
(primarily hub and O/D, but also seasonality, etc.), airside elements design (taxiway 
system and apron taxiway/taxilane system), etc.  
In this thesis the emphasis is on the difference between traffic patterns at O/D and hub 
airports i.e. between traffic distribution (“uniform”, concentrated in waves or 
combination of the two). In the general case (O/D airports) capacities provided by the 
apron and the runway system can be calculated independently and compared to each 
other to identify the bottleneck in the system. This is not the case for hub airports that 
serve coordinated flights32 concentrated in waves (solely, or in combination with other 
point-to-point flights33). The apron capacity depends on the capacity provided by the 
runway system. These two elements should be coupled with each other. 
5.5.1 Origin-destination airports 
The matter discussed above, related to transformation of arrivals/h to operations/h or 
vice versa, given a share of arrivals and departures in representative peak period, can 
apply to O/D airports. Consequently, provided the capacities of the runway system and 
apron(s) for typical peak periods, it is simple to identify which of the resources is the 
bottleneck, and to what extent the other resource is utilized.  
                                                 
32 The term “coordinated” aircraft/flight/traffic hereinafter refers to aircrfat carrying primarily transfer 
passengers. Coordinated aircraft are concentrated in waves, aimed at providing efficient transfers between 
flights.  
33 The term “point-to-point” flight/traffic hereinafter referrers to non-coordinated aircraft carrying origin-
destination passengers rather than transfer passengers.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
76 
 
Example 3: Let us observe a single-runway airport with 22 contact stands (of which 12 
for aircraft class 2, and 10 for aircraft class 3) and 8 remote stands (of which 5 for 
aircraft class 1, and 3 for aircraft class 2).  
Minimum ATC separation requirements between two consecutive arriving aircraft on 
approach are given in Table 9 (in Nm). Aircraft approach speeds are 110kts for class 1, 
130kts for class 2, and 150kts for class 3. Runway occupancy time is 40sec for all 
aircraft. Critical separation is imposed by the separation requirements in the air, 
implying that the runway occupancy time is shorter than the required separation time 
between any pair of the arriving aircraft.  
Minimum ATC separation requirements between two consecutive departing aircraft are 
given in Table 10 (in seconds). 
Table 9. Minimum ATC separation rules between arriving aircraft on approach in Example 3 
 
Table 10. Minimum ATC separation rules between departing aircraft in Example 3 
 
In addition, separation of 2Nm has to be provided between departing aircraft and 
arriving aircraft on approach. The length of the final approach path is assumed to be 
5Nm. 
All stands are available to all users with respect to airline/type of service i.e. there are 
no user constraints on stand usage. Turnaround times are 30min for class 1, 45min for 
class 2, and 60min for class 3 aircraft. Separation time between any two consecutive 
stand occupancies is assumed to be 5min.  
1 2 3
1 3 3 3
2 4 3 3
3 5 4 4
dij (Nm)
class j
cl
as
s 
i
1 2 3
1 45 45 45
2 60 60 60
3 120 90 90c
la
ss
 i
qij (s)
class j
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
77 
 
Aircraft fleet mix is: 20% class 1, 60% class 2 and 20% class 3 aircraft.  
Apron configuration (contact stands + remote stands), demand structure, and turnaround 
times and separation times are summarized in Table 11. 
Table 11. Apron configuration, demand structure, turnaround times and separation times in Example 3 
   
Based on the given inputs, single runway capacity envelope is derived and shown in 
Figure 26.  
 
Figure 26. Runway capacity envelope in Example 3 
Capacity of the runway is: 35 arrivals/h (plus additional 11 departures/h), 56 
departures/h and 66 operations/h in mix mode. 
Apron capacity is estimated as follows: 
( ) ⎟⎟⎠
⎞
⎜⎜⎝
⎛==
'
'minmin
i
i
iii t
NCC  
number of 
stands aircraft class
share in 
demand (%)
average turnaround 
time (min)
separation 
time (min)
        0+5      1 20 30 5
      12+3 2 60 45 5
      10+0 3 20 60 5
35; 0
35; 11
33; 33
0; 56
0
10
20
30
40
50
60
0 10 20 30 40 50 60
de
pa
rtu
re
s/
h
arrivals/h
single runway capacity envelope
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
78 
 
( ) ⎟⎠⎞⎜⎝⎛ ⋅⋅
+⋅⋅+⋅
+++⋅⋅+⋅+⋅
+++++== 60
652.0
010,60
652.0506.0
031012,60
652.0506.0352.0
03510120min,,min 3,21 CCCC
 
haircraftCC /9.342 ==  
Translated into operations/h, it makes 70 operations/h, assuming a 50/50% share of 
arrivals and departures in peak period. 
Capacity provided by contact stands only, i.e. gate capacity is: 
( ) ⎟⎟⎠
⎞
⎜⎜⎝
⎛==
'
'minmin
i
i
iii t
NCC  
( ) ⎟⎠⎞⎜⎝⎛ ⋅⋅⋅⋅+⋅
+⋅⋅+⋅+⋅
++== 60
652.0
10,60
652.0506.0
1012,60
652.0506.0352.0
10120min,,min 3,21 CCCC
 
haircraftCC /4.261 ==  
It makes 53 operations/h, assuming a 50/50% share of arrivals and departures in peak 
period. 
If all available stands are used, capacities provided by apron and runway are similar. 
The bottleneck is on the runway, which provides 93% utilization of apron capacity. 
If only contact stands are used (e.g. remote stands are in the maintenance area), then the 
bottleneck is the apron. Runway (mixed mode) capacity is engaged up to 81%, 
assuming 50/50% share of arrivals and departures in peak period34.  
5.5.2 Hub airports – only coordinated flights 
At hub airports relation between apron and the runway system is more complex than at 
O/D airports. In order to show such a relation, let us observe an ideal example of the 
hub airport where base airline/alliance operates exclusively waves of incoming and 
outgoing flights. The main parameters describing traffic pattern at hub airport i.e. wave-
system structure, are shown in Figure 27. 
                                                 
34 The utilization level is simple to calculate for other shares of arrivals and departures, depending on their 
mix in peak periods. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
79 
 
In this case, aircraft arrive during arrival time window (ARRtw), allow for at least 
minimum connecting time (MCT) for transfers between flights and then depart during 
departure time window (DEPtw). This is considered a wave.  
The wave length (WL) is determined as: 
twtw DEPMCTARRWL ++=                                                           (15) 
 
Figure 27. Parameters describing wave-system structure in the case of split waves (WRC≥WL) 
The wave length depends on the runway system performance, because it has direct 
impact on the lengths of the arrival and departure time windows. 
When discussing ideal wave of flights, the departure time window (DEPtw) is 
approximated to be of the same length as arrival time window (ARRtw), (Dennis, 1994; 
Burghouwt and de Wit, 2005; Burghouwt, 2007). This approximation is also adopted in 
the thesis. It makes wave length as: 
MCTARRWL tw +⋅= 2                                                                                   (16) 
Dependence of the wave length on changes in the number of aircraft per wave and MCT 
is shown in Figure 28. Runway capacity is assumed to be 30 arrivals or departures per 
hour (implying 2 minutes separation time). 
Wave length should satisfy the given level of service defined through the maximum 
acceptable connecting time (MaxCT) i.e. WL≤MaxCT. This means that the maximum 
number of aircraft per wave is limited either by apron static capacity or by MaxCT, 
whichever is more constraining.  
ARRtw MCT DEPtw
WL
WRC
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
80 
 
 
Figure 28. Dependence of the wave length on the number of aircraft per wave and MCT, assuming 
runway capacity of 30 arrivals/h 
The static capacity of an apron is the maximum simultaneous number of aircraft that can 
be accommodated at the aircraft stands on the apron. This capacity depends both on the 
apron configuration and demand structure. Assuming that single (home base) airline or 
alliance participates in the wave35, the static apron capacity36 is determined as follows: 
( ) ⎟⎟⎠
⎞
⎜⎜⎝
⎛==
'
'
minmin/
ij
ij
ijijijca s
N
NN
                                                                      (17) 
where: 
ijN - maximum simultaneous number of aircraft at the apron, limited by the 
group of stands available for user i and aircraft class j 
'ijN - number of stands that may be used by aircraft of user i and class j (stands 
allowed to be used by user i, designed for aircraft class j and for aircraft larger 
than j), see expression (11) 
                                                 
35 In this case, users can be defined with respect to type of service related to the level of security required 
inside the terminal building. 
36 It is derived from the condition that the number of stands supplied has to be larger or equal to the 
number of stands demanded: ijijij NsN ⋅≥ '' , analogue to expression (13) for deriving dynamic capacity. 
0
50
100
150
200
250
300
350
5 10 15 20 25 30 35 40 45 50
w
av
e 
le
ng
th
 (m
in
)
no. of aircraft in a wave
MCT=30 MCT=40 MCT=50 MCT=60 MCT=90
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
81 
 
'ijs  - cumulative share of user/aircraft class combination allowed to use the ij
th 
group of stands 
∑∑
∈ ∈
=
Kk Ll
klij ps '                 (18) 
{kK = [ ]nk ,1∈  and user-class k allows its stands to be used by user-class }i  
{lL = [ ]mj ,1∈  and aircraft class l is equal or larger than aircraft class j, 
}jl ≥  
klp  – proportion of aircraft of user k and class l in the population of aircraft 
demanding service 
If the level of service is the limiting factor, it follows that: 
MaxCTWL =            (19) 
MCTARRMaxCT tw +⋅= 2                                        (20)   
Alostw tNARRMCTMaxCT ⋅⋅=⋅=− 22         (21) 
⇒
A
los t
MCTMaxCTN ⋅
−=
2              (22) 
where: 
At - weighted average separation time of all arriving aircraft on the runway37. 
Dependence of Nlos on changes in MaxCT and MCT is shown in Figure 29. Runway 
capacity is again assumed to be 30 arrivals or departures per hour (implying 2 minutes 
separation time). 
                                                 
37 In the general case, expression (22) would be: 
DA
tt
MCTMaxCT
losN +
−= , where: Dt - is weighted average 
separation time between all departing aircraft on the runway.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
82 
 
 
Figure 29. Dependance of Nlos on MaxCT and MCT, assuming runway capacity of 30 arrivals/h 
Maximum number of aircraft that can be scheduled within a wave is: 
( )losca NNN ,min /=                                                                                      (23) 
Depending on which one is more constraining, the maximum number of aircraft is a 
function of runway performance (if it is limited by Nlos) or this effect is omitted (if static 
apron capacity is more critical). 
Assuming that arrival and departure waves are of the same length and the sequence of 
arrivals is the same as the sequence of departures, the average turnaround time (TAT) of 
the aircraft connected in the wave can be estimated as it follows: 
MCTtNMCTARRTAT Atw +⋅=+=                                                         (24) 
The parameter that describes the demand pattern is the wave repeat cycle (WRC). It is 
the time interval between the same points of the two consecutive waves.  
Figure 27 has shown the case with split waves (WRC≥WL). Otherwise, if WRC<WL, 
waves partially overlap (departure time window with arrival time window of the next 
wave) as it is given in Figure 30. 
0
10
20
30
40
50
60
70
80
90 120 150 180 210 240 270 300 330
N
lo
s (
ai
rc
ra
ft
)
MaxCT (min)
MCT=30 MCT=40 MCT=50 MCT=60 MCT=90
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
83 
 
 
 
Figure 30. Parameters describing wave-system structure in the case of overlapping waves, WRC<WL 
The theoretical capacity assumes exchange of aircraft i.e. new aircraft arrive as soon as 
the previous aircraft release stands. When an airline concentrates its flights in waves, 
the theoretical capacity is reached in the case with complete overlapping between the 
departure time window and the following arrival time window, as shown in Figure 31, 
i.e. SOTWRC = .  
 
Figure 31. Parameters describing wave-system structure in the case with entirely overlapped waves, 
WRC=SOT 
“Providing” exchange of arrivals and departures on the runway requires38 the following: 
p
Atwtwtw tNARRDEPDEPARR ⋅===                                                   (25) 
where: 
p
At  - weighted average separation time of all arriving aircraft on the runway 
allowing one departure to be performed between any two consecutive arrivals 
                                                 
38  At airports with parallel runways, one of which is for arrival and one is for departure, ARRtw is equal 
to AtN ⋅ . 
WL
WRC
ARRDEPtw MCT ARRDEPtw
WRC
WL
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
84 
 
Assuming an exchange of aircraft on the runway and apron, maximum number of 
aircraft per wave limited by the level of service is: 
MCTARRDEPMaxCT tw +⋅= 2                                                           (26) 
p
Alostw tNARRDEPMCTMaxCT ⋅⋅=⋅=− 22         (27) 
⇒ p
A
los
t
MCTMaxCTN ⋅
−=
2                                                                            
(28) 
With the number of aircraft per wave N (calculated from expression 23), the average 
stand occupancy time can be calculated as it follows: 
STMCTtNSTMCTARRDEPSTTATSOT pAtw ++⋅=++=+=     (29) 
Consequently, the theoretical apron capacity for serving coordinated flights is39: 
SOT
NCT =                                                                                                       (30) 
Theoretical apron capacity to serve waves of flights is exploited only to a certain extent. 
If no other flights are scheduled until the next wave, WRC can be observed as certain 
form of stand blocking time. This means that the utilized capacity (due to demand 
requirement) cannot be more than: 
WRC
NCU =                                                                                                      (31) 
Utilization of theoretical capacity by the given demand is: 
T
U
C
CU = %                                                                                                       (32) 
                                                 
39 In the case in which the number of aircraft per wave (N) is limited by the static apron capacity (Na/c), 
expression (30) could be derived also from the general apron capacity estimation model (Section 5.1), 
because static capacity already takes into account shares in demand, and SOTs for all users are equal. The 
same does not apply when Nlos imposes the constraint on the number of aircraft per wave. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
85 
 
In the case with non-overlapping waves WRC≥WL the airline effectively uses runway 
capacity during ARRtw and DEPtw, while during the MCT period and WRC-WL period 
it remains idle.  
In the case of completely overlapping waves, periods when the runway is effectively 
used in mix mode (ARRDEPtw) alternate with idle MCT periods.  
The entire discussion, as given above, refers to wave of flights having all flights 
connected to each other (e.g. continental connections). In the case of mixed 
intercontinental (IC) and continental (C) connections, the calculation of Nlos and TAT 
for different groups of flights (with respect to type of service) is given in Appendix 12. 
 
Example 3a: Let us observe the same airport as in Example 3, i.e. single-runway airport 
with 22 contact stands (of which 12 for aircraft class 2, and 10 for aircraft class 3) and 8 
remote stands (of which 5 for aircraft class 1, and 3 for aircraft class 2). The base airline 
operates waves of flights with WRC of 180min. It uses only contact stands to provide a 
MCT of 30min. Remote stands are used only for aircraft overnights. MaxCT is 150min. 
Traffic structure is 20% aircraft class 1, 60% aircraft class 2, and 20% aircraft class 3 
(as it is in Example 3). 
Static apron capacity is determined as it follows: 
⎟⎟⎠
⎞
⎜⎜⎝
⎛=
'
'min/
i
i
ica s
NN
 
( ) ⎟⎠⎞⎜⎝⎛ +
+
++
++==
2.0
10,
2.06.0
1012,
2.06.02.0
10120minmin 3,2,1/ NNNN ca
 
aircraftNN ca 221/ ==  
In the case of split waves, the maximum number of aircraft that can be scheduled in a 
wave, to satisfy the level of service, would be: 
aircraft
t
MCTMaxCTN
A
los 3569.12
30150
2
=⋅
−=⋅
−=
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
86 
 
As a reference for theoretical capacity, complete overlapping has to be considered, 
implying: 
aircraft
t
MCTMaxCTN
p
A
los 3284.12
30150
2
=⋅
−=⋅
−=  
In both cases apron static capacity appears to be more critical.  
( ) aircraftNNN losca 22,min / ==
 
Stand occupancy time is determined as: 
min5.7553084.122 =++⋅=++⋅= STMCTtNSOT p  
Consequently, the theoretical apron capacity is determined as: 
haircraft
SOT
NC /5.17==
 
The utilized capacity with respect to demand requirements i.e. wave-system structure, 
is: 
haircraft
WRC
NCU /3.7==  
It implies that the base airline uses 42% of theoretical apron capacity: 
42.0===
WRC
SOT
C
CU
T
U   
Figure 32 shows the theoretical and utilized apron capacity for different wave-system 
structures i.e. WRCs. All inputs are the same as in Example 3a, only WRC varies from 
300min to SOT. It can be seen that theoretical capacity does not change, because SOT 
remains the same, regardless of WRC. As WRC decreases, utilized capacity increases, 
until it reaches the theoretical capacity, when WRC becomes equal to SOT. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
87 
 
 
Figure 32. Theoretical vs. utilized capacity for different WRCs in Example 3a 
Figure 33 shows how maximum number of aircraft per wave changes with MaxCT40. 
Under the conditions observed in Example 3a, MaxCT below 110min would become 
constraint for the maximum number of aircraft per wave. 
 
Figure 33. Maximum number of aircraft per wave for different MaxCTs in Example 3a 
                                                 
40 Nlos is calculated using 
p
At .  
0
2
4
6
8
10
12
14
16
18
20
60 80 100 120 140 160 180 200 220 240 260 280 300
ap
ro
n 
ca
pa
ci
ty
 (a
ir
cr
af
t/
h)
WRC (min)
utilized capacity theoretical capacity
0
5
10
15
20
25
30
35
40
45
90 100 110 120 130 140 150 160 170 180
nu
m
be
r o
f a
ir
cr
fa
t
MaxCT (min)
Na/c Nlos N
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
88 
 
In the case of split waves, 22 aircraft are served during the period of wave length: 
min6.1042 =+⋅= MCTARRWL tw  
The runway is effectively used for arrivals during 37min, followed by idle period of 
30min, before it is used by departures during another 37min. 
The remaining period until the next wave (WRC-WL) of 75.4min remains idle. 
Literature review showed that only Steuart (1974) and Hassounah and Stuart (1993) 
addressed the issue of impact of wave (bank) operations on requirements for contact 
stands (gate) positions. They suggested how to calculate the expected number of aircraft 
occupying gate positions at time t from a given schedule, accounting for random 
deviations from the schedule. Steuart (1974) observed the scheduling alternatives to 
decrease (increase) the number of aircraft in the bank, while decreasing (increasing) the 
time interval between the banks (i.e. WRC), under constant arrival rate. He analyzed the 
changes in the expected number of aircraft at gates for short time intervals, from 0min 
(random schedule) to 60min, which are more of a theoretical approach than reflection of 
the real scheduling practice.  
Hassounah and Steuart (1993) briefly discussed an impact of the time interval between 
the banks on gate requirements on numerical example, given a fixed number of aircraft 
per wave and fixed stand occupancy time. They analyzed more realistic time intervals, 
from 60min to 150min. They showed, as the spacing between the two banks decreases, 
the required number of gates starts to increase, reaching its maximum when the 
departure of the flights of the first bank and the arrival of the flights of the second bank 
coincide. Authors did not observe influence of the number of coordinated aircraft in the 
bank on average stand occupancy time, nor the influence of the runway system 
performance on apron capacity.   
In addition, let us also consider the case when all 30 stands (contact + remote) can be 
engaged for coordinated flights, solely for the purpose of comparison to O/D airport. In 
this case, the maximum number of aircraft per wave is: 
( ) aircraftNNN losca 30,min / ==  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
89 
 
The level of service becomes a greater constraint than static capacity only when 
MaxCT≤140min (see Figure 34).  
 
Figure 34. Maximum number of aircraft per wave for different MaxCTs in Example 3a (contact and 
remote stands) 
In this case, stand occupancy time changes. It is: 
min2.90=SOT  
Consequently, theoretical and utilized capacities are: 
haircraftCT /9.19=  and  
haircraftCU /10= , respectively. 
Figure 35 shows the capacity of an apron with 22 contact and 8 remotes stands for O/D 
airport case (Example 3) and pure hub airport serving only coordinated flights (Example 
3a). The dark blue bar represents the capacity of contact stands, while the light blue bar 
represents additional capacity if remote stands are also engaged.  
 
0
5
10
15
20
25
30
35
40
45
90 100 110 120 130 140 150 160 170 180
nu
m
be
r o
f a
ir
cr
fa
t
MaxCT (min)
Na/c Nlos N
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
90 
 
 
Figure 35. Apron capacities for O/D and hub airports (Examples 3 and 3a) 
The capacity of the same apron serving the same fleet mix is as much as 50% higher in 
the case of O/D airport in comparison to hub. If all stands are taken into consideration41 
this difference increases to 65%. The reason lies in the significant difference in 
weighted stand occupancy times. Theoretical capacity for hub airport is used as a 
reference for comparison, because the utilized capacity is a reflection of demand pattern 
(WRC=180min), not the system’s ability to handle it. 
The blue line represents the arrival capacity provided by the runway operating in mix 
mode. As discussed in Example 3, it is lower than apron capacity when all stands are 
used, but exploited to 83% when only contact stands are considered. In the case of hub 
airport there is not much sense to compare runway and apron capacities. In order to 
provide greater number and better quality of connections, the runway is exploited to 
100% in the periods of ARR and DEP time windows while the rest of time it remains 
idle. 
 
                                                 
41 The percentage is expressed relative to runway capacity in this case, because it does not allow more 
than 33 arrivals/h in mixed mode, which is less of the apron potential to serve 35 aircraft/h. 
26
17
8
2
0
5
10
15
20
25
30
35
O/D  HUB ‐ theoretical
ap
ro
n 
ca
pa
ci
ty
 (a
ir
cr
af
t/
h)
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
91 
 
5.5.3 Hub airports – mix of coordinated and point-to-point flights 
If static apron capacity is greater than the number of aircraft per wave constrained by 
MaxCT, it could theoretically happen that: WRC<ARRtw+MCT. However, it is not 
reasonable for this to happen in practice. In such circumstances, base airline can rather 
decide to expand the waves (increase the number of aircraft in a wave), decreasing the 
level of service to some extent. The other possibility is that the remaining apron 
capacity is used by other airlines operating point-to-point flights (in strong O/D 
markets), i.e. not participating in the wave-system.  
Additional point-to-point flights are even more reasonable to expect in off-wave 
periods, which begin when connected aircraft, after exchanging passengers, start to 
leave the apron, and last until the next arrival wave.  
In the case of airports with mixed coordinated and point-to-point traffic, in addition to 
airline and/or type of service/security level, different users (“customers”) have to be 
defined also with respect to the nature of traffic. It is because different stand occupancy 
times apply to flights/aircraft coordinated flights and to point-to-point flights. The 
period of analysis should be the wave repeat cycle. 
Apron capacity is determined as the minimum of the capacities set by the group of 
stands for different types of traffic, implying the following: 
 
( )III CCC ,min=                   (33) 
where: 
IC - apron capacity limited by the group of stands available for coordinated 
flights 
IIC - apron capacity limited by the group of stands available for other point-to-
point (non-coordinated) flights 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
92 
 
Let us assume that out of n users, [ ]ni ,1∈ , the first g users belong to group of 
coordinated flights, [ ]gq ,1∈ , and the rest (n-g) belong to group point-to-point flights, 
[ ]ngr ,1+∈ . 
For the group of stands available for coordinated traffic, apron capacity (CI) is 
calculated using expression (31), having a maximum number of stands per wave 
determined according to expression (23) (based on expressions (17) and (28)). These 
expressions have to be corrected for the share of coordinated traffic in total demand (sI), 
during period observed (WRC), as given below: 
II
U
I sWRC
N
s
CC ⋅==                                                                                      (34) 
where:  
∑∑
= =
=
g
q
m
j
qjI ps
1 1            
(35) 
q designates the user that belongs to the group of coordinated flights, 
[ ]gq ,1∈  
j designates the aircraft class, [ ]mj ,1∈ , where 1 stands for the smallest 
aircraft class, and m for the largest aircraft class 
The maximum number of aircraft per wave is determined using expression (23), as 
follows: 
( )losca NNN ,min /=  
where the maximum number of aircraft constrained by the level of service is determined 
using expression (28): 
p
A
los
t
MCTMaxCTN ⋅
−=
2
 
while the static apron capacity is determined using the modified expression (17), as 
follows: 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
93 
 
⎟⎟⎠
⎞
⎜⎜⎝
⎛ ⋅= I
qj
qj
qjca
s
s
N
N
'
'
min/        (36) 
∑∑
∈ ∈
=
Oo Ll
olqj NN '  42         (37) 
∑∑
∈ ∈
=
Oo Ll
olqj ps '         (38) 
{oO = [ ]go ,1∈  and user-class o allow its stands to be used by user-class 
}q , [ ]gO ,1⊆   
{lL = [ ]mj ,1∈  and aircraft class l is equal or larger than aircraft class j, 
}jl ≥  
For the group of stands available for point-to-point traffic, apron capacity (CII) is 
calculated using expression (10). Only those sub-groups that are related to users r, 
[ ]ngr ,1+∈  are considered. Corresponding shares, with respect to total demand during 
the observed period, are taken into account. The following is implied: 
( ) ⎟⎟⎠
⎞
⎜⎜⎝
⎛==
'
'
minmin
rj
rj
rjrjrj t
N
CC
       
(39) 
where: 
r designates the user that belongs to the group of other (non-coordinated) flights, 
[ ]ngr ,1+∈   
j designates the aircraft class, [ ]mj ,1∈ , where 1 stands for the smallest aircraft 
class, and m for the largest aircraft class 
                                                 
42 If remote stands for other users can also be used by coordinated flights it becomes: ∑∑
∈ ∈
=
Kk Ll
klqj NN ' ;
{kK = [ ]nk ,1∈  and  user-class k allows its stands to be used by user-class }q . This is applicable only for 
preferential use case. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
94 
 
The number of stands for each sub-group of stands and corresponding weighted average 
stand occupancy times are determined as in the basic case, using expressions (11) and 
(12), as follows: 
∑∑
∈ ∈
=
Kk Ll
klrj NN '
 
∑∑
∈ ∈
⋅=
Kk Ll
klklrj Tpt '
 
{kK = [ ]nk ,1∈  and user-class k allows its stands to be used by user-class 
}r , [ ]nK ,1⊆  
{lL = [ ]mj ,1∈  and aircraft class l is equal or larger than aircraft class j, 
}jl ≥   
Let us observe the case where in addition to coordinated flights, the same or other 
airlines operate also other point-to-point flights (low-costs, regionals, charters, etc.). 
They can use either remote stands only (exclusive use case), or they are also allowed to 
use contact stands during off-wave periods (preferential use case).  
Example 3b: Apron configuration and runway parameters are the same as in Example 3: 
single-runway airport with 22 contact stands (of which 12 for aircraft class 2, and 10 for 
aircraft class 3) and 8 remote stands (of which 5 for aircraft class 1, and 3 for aircraft 
class 2).  Hubbing parameters are the same as in Example 3a: WRC is 180min; MCT is 
30min and MaxCT is 150min.  
During the WRC period demand structure is 40% (sI) coordinated flights (of which 20% 
aircraft class 1, 60% aircraft class 2, and 20% aircraft 3) and 60% other flights (of 
which 60% aircraft class 1, and 40% aircraft class 2 aircraft). Table 12 summarizes 
information on apron configuration, demand structure during the WRC period, average 
turnaround times and separation time. 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
95 
 
Table 12. Apron configuration, demand structure, turnaround times and separation times in Example 3b 
 
Two cases are analyzed: 
1. Exclusive use case: contact stands are exclusively used by coordinated flights; 
remote stands are available for other flights; and  
2. Preferential use case: contact stands are also available for other (point-to-point) 
flights when they are not used by coordinated flights. 
Case 1 – exclusive use 
For the exclusive use case contact stands are blocked for use by other users, implying 
that no one else, besides coordinated flights can use them during the WRC period. Other 
aircraft can exchange only on remote stands after being turned-around and separation 
from other aircraft is provided. Overall apron capacity is determined as the minimum of 
the capacity limited by the group of stands for coordinated flights i.e. contact stands (CI) 
and the capacity limited by the group of stands for other flights i.e. remote stands (CII), 
under the given demand structure. It follows43: 
4.0
3.7
4.0
=== WRC
N
s
C
C
I
U
I
    
 
haircraftCI /3.18=
 ( )
( )15,5.19min60
5024.0
3,60
5024.03536.0
35min
,min 21
=⎟⎠
⎞⎜⎝
⎛ ⋅⋅⋅⋅+⋅
+=
=
II
IIIIII
C
CCC
 
                                                 
43 CU is calculates as in Example 3, expression (31). N is derived from expression (23), where Na/c from 
(17) and Nlos from (28); sI is the share of coordinated flights (0.4 in this example). 
number of stands traffic type aircraft class share in demand (%)
average turnaround 
time (min)
separation 
time (min)
0 1 8 70,5 5
12 2 24 70,5 5
10 3 8 70,5 5
5 1 36 30 5
3 2 24 45 5
coordinated
other
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
96 
 
haircraftCII /15=  
( ) haircraftCCC III /15,min ==  
Apron capacity in the exclusive use case is 15 aircraft/h and is limited by the group of 
stands for other flights. 
Case 2 – preferential use 
In the preferential use case contact stands are used by coordinated flights, but when they 
leave the stands (after ARRtw+MCT+ST), until the next wave (WRC), they can be used 
by other aircraft, in addition to remote stands. Overall apron capacity is the minimum of 
the capacity limited by the group of stands for coordinated flights (CI) and the capacity 
limited by the group of stands for other flights (CII), under the given demand structure. 
Capacity CI is calculated in the same way as in Case 1, since the next wave of 
coordinated flights does not arrive before WRC. Capacity CII takes into account that 
contact stands are released after the period: ARRtw+MCT+ST. It follows: 
4.0
3.7
4.0
== UI CC
 
haircraftCI /3.18=  
( )21,min IIIIII CCC =  
⎟⎟⎠
⎞
⎜⎜⎝
⎛ ⋅⋅+⋅+
++⋅⋅+⋅+⋅++
++++= 60
5024.05.75)08.024.0(
31012,60
5024.03536.05.75)08.024.008.0(
3510120minIIC
 
( )5.42,8.32min=IIC  
haircraftCII /8.32=  
( ) haircraftCCC III /3.18,min ==
 
Apron capacity in the preferential use case is 18.3 aircraft/h and is limited by the group 
of stands for coordinated flights.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
97 
 
In the exclusive use scenario contact stands are utilized only to 42% (as shown in 
Example 3a) by the coordinated flights. By allowing them to be used by point-to-point 
flights between consecutive waves in the preferential use scenario, it results in a 22% 
higher capacity. 
Example 4 is used to show apron capacity changes in the preferential and exclusive use 
cases, with different shares of coordinated flights and other flights during the WRC 
period. 
Example 4: The same airport is observed as in Example 3: single-runway airport with 22 
contact stands (of which 12 for aircraft class 2, and 10 for aircraft class 3) and 8 remote 
stands (of which 5 for aircraft class 1, and 3 for aircraft class 2).  Hubbing parameters 
are the same as in Example 3a: WRC is 180min; MCT is 30min and MaxCT is 150min. 
The average turnaround time for other, non-coordinated aircraft, operating point-to-
point flights is 40min. Runway capacity is the same as in Example 3. 
During the WRC period the demand structure, in terms of share coordinated flights vs. 
other flights, is varied from 0/100 % to 100/0 %, as shown in Figure 36. All the other 
physical and operational system parameters are assumed as in Example 3b.  
 
Figure 36. Apron capacity sensitivity to share of coordinated and other flights under exclusive and 
preferential use cases in Example 4 
0
5
10
15
20
25
30
35
40
45
0 10 20 30 40 50 60 70 80 90 100
ap
ro
n 
cp
ac
ity
 (a
ir
cr
af
t/
h)
 
share of coordinated flights (%)
exclusive use preferential use
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
98 
 
It can be seen that once the constraint moves from the contact stands area, designated 
for coordinated flights, to the group of stands available for other flights, preferential use 
case provides higher capacity figures than exclusive use case.  
In Figure 37 the same relation is shown, only expressed in the number of coordinated 
and other aircraft that can be served during the period of 3h (WRC). The corresponding 
values are summarized in Table 13. 
 
Figure 37. Relationship between number of coordinated and other aircraft handled on the apron, during 
the WRC, for different demand structures, under exclusive and preferential use cases in Example 4 
It can be seen that the number of coordinated flights per wave does not exceed the limit 
of 22 aircraft (set by contact stands static capacity). The number of other flights that can 
be served during WRC varies depending on their share in demand and the policy of 
contact stand use. 
In his work Steuart (1974) addressed the impact of different scheduling practices on 
gate requirements, including wave operations and the difference between exclusive and 
common use gate strategies. However, he observed these two independently one of the 
other.   
0
20
40
60
80
100
120
140
0 5 10 15 20 25
ot
he
r f
lig
ht
s 
pe
r W
RC
coordinated flights per WRC
preferential use exclusive use
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
99 
 
Table 13. Estimated number of coordinated and other aircraft handled on the apron, during the WRC, for 
different demand structures, under exclusive and preferential use cases in Example 4 
 
 
Figure 38 summarizes apron capacities for pure hub (Example 3a) and for hub with 
additional point-to-point traffic (Example 3b).  
 
Figure 38. Apron capacities for different airport types (Examples 3a and 3b and modified Example 3)  
The aim is to show apron capacity changes at hub airports when additional point-to-
point traffic is allowed to use only remote stands (exclusive use), and contact stands in 
addition, during off-wave periods (preferential use). Furthermore, the O/D airport case  
coordinated other coordinated other
0/100 0 32 0 120
10/90 4 32 11 101
20/80 8 32 21 85
30/70 14 32 22 51
40/60 21 32 22 33
50/50 22 22 22 22
60/40 22 15 22 15
70/30 22 9 22 9
80/20 22 6 22 6
90/10 22 2 22 2
100/0 22 0 22 0
exclusive use preferential usecoordinated/
other flights 
share
7
15
18
30
11
0
5
10
15
20
25
30
35
40
45
HUB ‐ utilized MIX ‐ exclusive MIX ‐
preferential
O/D 
ap
ro
n 
ca
pa
ci
ty
 (a
ir
cr
af
t/
h)
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
100 
 
is given as a reflection of the capacity if only point-to-point traffic is served at all 
stands. 
For the case of pure hub, the comparison reference is the contact stand utilized capacity 
(with respect to WRC), because coordinated flights also use only contact stands in the 
exclusive/preferential use cases. The same fleet mix as used for the pure hub (Example 
3a) is used for coordinated flights in exclusive/preferential cases (Example 3b).  
In order to make the O/D case comparable to hub (mainly preferential) case, the overall 
fleet mix at exclusive/preferential case (Example 3b) is also applied for the O/D case. 
Table 14 summarizes the demand structure for different cases compared in Figure 38. 
In this particular case (40/60 share of coordinated/point-to-point flights) apron capacity 
is two times higher (7.3 vs. 15 aircraft/h) when only (eight) remote stands are engaged 
for point-to-point traffic (exclusive use) and as much as 2.5 times higher (7.3 vs. 18.3 
aircraft/h) when point-to-point traffic can also use (22) contact stands during off-wave 
periods, in addition to remote stands (preferential use).  
Table 14. Demand structure for pure hub (Example 3a), for hub with additional point-to-point traffic 
(Example 3b) and for O/D airport (modified Example 3) 
 
With only point-to-point traffic served (O/D case), runway capacity becomes the 
constraining factor. It is 78% higher than apron capacity in the preferential use case. 
There is no sense comparing O/D to the exclusive use case, since contact stands are not 
used in the same regime. 
flight type aircraft class HUB pure
HUB 
exclusive/
preferential
O/D
1 0,2 0,08 0
2 0,6 0,24 0
3 0,2 0,08 0
1 0 0,36 0,44
2 0 0,24 0,48
3 0 0 0,08
coordinated
other
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
101 
 
5.5.4 The impact of runway system performance on apron capacity 
As discussed earlier, the runway system capacity does not have any impact on apron 
capacity for O/D airports. Apron and the runway system capacities are calculated 
separately, based on the variables that each of them depends on. The one that allows a 
lower capacity imposes the constraint on the overall airside capacity.  
At hub airports apron and the runway system capacities cannot be observed separately. 
Capacity of apron to handle coordinated flights is affected by runway system 
performance characteristics.  
Maximum number of aircraft within the wave may depend on apron configuration and 
demand structure (static apron capacity), or it may be a function of runway system 
performance (through maximum acceptable connecting time). If the number of aircraft 
per wave is limited by the level of service, i.e. not by static apron capacity, this would 
make apron capacities (both theoretical and utilized) dependant on the runway system 
performance. 
If static capacity sets the limit for maximum number of aircraft per wave, then: 
‐ Theoretical apron capacity still depends on the runway system performance 
because SOT is a  function of weighted average separation times on the runway; 
‐ Utilized apron capacity does not depend on runway performance, since being 
derived from WRC, which is a characteristic of the demand itself, not a 
reflection of runway performance. 
Example 5 assumes different minimum separation rules between arriving and departing 
aircraft on the runway compared to Example 3. The minimum separation between all 
aircraft classes on approach is 5Nm, and minimum separation between all departing 
aircraft is 120sec. The apron has the same structure as in Example 3, which is: 22 
contact stands (of which 12 for aircraft class 2, and 10 for aircraft class 3) and 8 remote 
stands (of which 5 for aircraft class 1, and 3 for aircraft class 2). Hubbing parameters 
are the same as in Example 3a: WRC is 180min; MCT is 30min and MaxCT is 150min. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
102 
 
The demand structure is the same as given in Table 14 (modified O/D example is used 
as comparable to hub with mixed coordinated and point-to-point flights).  
In this case, runway capacity drops to 25 arrivals/h, 50 operations/h (alternating arrivals 
and departures), and 30 departures/h. The impact of runway performance on apron 
capacities is shown in Figure 39.  
 
Figure 39. Apron capacity sensitivity to runway system performance, comparison between Example 3 and 
Example 5 
The blue and red lines represent runway capacity (arrivals/h) in mix mode, for Example 
3 and Example 5, respectively. In both examples airside capacity is limited by runway 
capacity for the O/D case and by apron capacity for hub cases.  
In the O/D case apron capacity remains the same regardless of runway capacit, only the 
mismatch between the two is higher in Example 5 (40 vs. 25 arrivals/h) than in Example 
3 (40 vs. 33 arrivals/h). 
At hub cases, the influence of runway capacity on apron capacity is not clearly visible in 
all of them. This is due to fact that only some variables that determine apron capacity 
are functions of runway performance, as discussed earlier.  
The maximum number of aircraft scheduled in a wave of flights, limited by the 
maximum acceptable connecting time (LOS capacity i.e. Nlos), decreases from 32 to 24 
0
5
10
15
20
25
30
35
40
45
O/D ‐ all Static 
capacity
LOS 
capacity
HUB ‐
theoretical
HUB ‐
utilized
MIX ‐
exclusive
MIX ‐
preferential
ap
ro
n 
ca
pa
ci
ty
 (a
ir
cr
fa
t/
h)
Example 3 Examle 5
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
103 
 
aircraft per wave44. In both cases it is still higher than static apron capacity (Na/c=22 
aircraft), which means that in this case static apron capacity dictates the maximum 
number of coordinated aircraft in a wave. 
Due to this, utilized capacity is not influenced by changes in runway performance, being 
derived from static capacity and WRC, which reflects demand characteristics. The 
theoretical capacity of the hub to handle coordinated flights decreases, due to the 
increase in ARRtw, and SOT.  
The impact of runway performance on apron capacity at airports with mixed 
coordinated and point-to-point flights exists, but it may be hidden, like it is in the case 
shown in Figure 39. It results from the more significant influence of other variables that 
affect apron capacity. In the exclusive use case, apron capacity is limited (to 15 
aircraft/h) by the capacity of group of stands for other flights/aircraft class 2. In the 
preferential use case, capacity (of 18.3 aircraft/h) is constrained by the (utilized) 
capacity of the group of stand for coordinated flights (i.e. contact stands). The effect of 
runway parameters on apron capacity is not visible, because in these examples the 
constraining group of stands is not sensitive to them. 
The additional example is given in Appendix 13. It assumes 6Nm separation between all 
aircraft classes on approach. In this case, Nlos becomes more constraining than static 
capacity. The influence of runway performance on apron capacity is also visible in the 
pure-hub and preferential use case. 
5.6 Reserve apron capacity 
Forsyth (2007) explores the main problems that changing demand poses to airports and 
how they cope with changes in the level and pattern of demand. He emphasizes aviation 
trends in Europe and their implications on airports. Five main aspects are observed: 
demand growth in the short run and in the long run; impact of new business models 
(particularly LCCs) and new aircraft types (A380 and B787) on the demand/capacity 
                                                 
44 Nlos is calculated with 
p
At which is more constraining than when calculated using At .  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
104 
 
balance; as well as improving airport efficiency and reducing costs in a competitive 
environment.  
In order to cope with demand growth in the short run (when expansion of the airport 
airside elements is not an option), airport has to ensure that the existing capacity is 
utilized more efficiently. Two mechanisms applicable for avoiding congestion on the 
short run, under growing demand and unchanged infrastructure: slot allocation and 
congestion pricing (Forsyth, 2007). However, in the long run, airport needs to expand it 
capacity in accordance to projected demand growth (too little investment leads to 
congestion or suppressed demand; while too much investment leads to expensive use).  
Airport capacity expansion is usually connected to huge infrastructural investments, 
primarily building new runway and/or terminal complex (terminal building and 
apron(s)). In order to meet demand volume and structure changes between these 
significant capacity jumps, airport capacity may be increased in smaller steps mainly 
through expansion and/or modification of existing elements, but only to a certain extent.  
When it comes to runway system capacity scarcity, there is not much that can be done 
before building a new runway. Runway extension and widening enable larger aircraft to 
land and take-off, and may indirectly increase the overall airport throughput in terms of 
volume of accommodated passengers, while the runway system capacity may even 
decrease due to increase in separation rules. In the case of small airports, the runway 
system capacity may be increased with some structural changes to the taxiway system 
(such as introduction of high speed exits and parallel taxiways). Additional gain may be 
achieved through technological improvements i.e. better navigational equipment 
(category upgrade) to increase runway system utilization, and/or different tools for 
supporting managing (sequencing) the runway operations. Also, some operational 
measures may result in higher runway system capacity e.g. intersection take-off, 
staggered and/or steeper approach at parallel runways, etc.  
On the other hand, before terminal complex reaches its capacity limit (usually expressed 
in the number of passengers that may be served through the terminal building on an 
annual basis), it can be modified and extended in smaller steps, as a response to demand 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
105 
 
characteristics changes. It is in a way a more “live” and adaptable airport element than 
the runway system.  To increase terminal complex capacity, the airport may:  
‐ In the terminal building: expand the terminal building (corridors, waiting rooms, 
etc.), engage additional equipment or introduce new technologies to improve 
passenger processing (check-in, baggage claim, security, waiting rooms, etc.), 
install additional air bridges, etc.;  
‐ In the apron area: invest in fixed installations to improve efficiency of aircraft 
turnaround, rearrange the apron area to accept different aircraft classes and/or 
different users (with respect to different criteria), expand the apron area with 
additional stands and/or parallel taxiways, etc.  
One of the above mentioned modifications related to apron area relate to the concept of 
reserve capacity (Mirkovic, 2011b and Mirkovic and Tosic, 2013). This thesis analyses 
apron potential to gain additional capacity (prior to physical expansion) through 
appropriate reallocation of available resources, mainly by changing i.e. relaxing existing 
operational constraints.  
Representation of apron capacity as given in Section 5.3 allows the comparison of the 
effects resulting from the changes on the supply side in order to meet expected demand 
changes. However, this type of envelope does not provide information about the extent 
to which the apron is already utilized by the traffic it serves. If the runway capacity 
envelope is presented using a scatter diagram (having dominant users on the x and y 
axes), this information becomes available.  
In Figure 40 the apron capacity envelope for Example 1 is given. It assumes: 5 stands 
for domestic and 6 stands for all flights; average stand occupancy times of 45min for 
domestic and 55min for other rotations. The number of aircraft flying on domestic 
rotation are given on the x-axis and the number of aircraft other than domestic rotations 
(purely international and mixed domestic-international) on the y-axis.  
The shares: 100/0, 80/20, 60/40, 50/50, 40/60, 20/80 and 0/100 of domestic/other 
rotations in total traffic are used to create the envelope. It shows expected, as the 
number of other rotations increases, the capacity decreases since the domestic stands are 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
106 
 
not available for other rotations. This apron configuration appears better adjusted to 
traffic with a greater share of domestic flights, as they can use all the stands at the 
apron. 
 
Figure 40. Apron capacity envelope for Example 1, scatter diagram 
Figure 41 depicts apron capacity envelope for apron at which groups of stands are 
strictly divided between the users, like in Example 2a (5 stands for the base airline and 6 
stands for all other airlines, assuming average stand occupancy times of 45min for the 
base airline and 55min for other airlines). The same shares as in the previous case are 
used (100/0, 80/20, 60/40, 50/50, 40/60, 20/80 and 0/100) to create apron capacity 
envelope. Only, in this case, these are shares of the base airline and other airlines in 
total traffic.  
Depending on the operational constraints applied, each apron configuration has the 
ability to respond to demand structure changes with respect to users, which is its apron 
capacity flexibility. But, it also has some reserve/latent capacity that may be activated 
by changing/relaxing user constraints. How much an apron can gain through relaxation 
of the constraints depends on the level of flexibility that is already built in with existing 
0
1
2
3
4
5
6
7
8
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
ot
he
r O
/D
 c
om
b.
 (a
irc
ra
ft
/h
)
domestic O/D comb. (aircraf t/h)
5 dom.f l. stands + 6 ALL f l. stands 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
107 
 
operational rules. Generally, fewer constraints (more shared resources) lead to higher 
flexibility, which implies that a constraint-free apron configuration is absolutely 
flexible.  
 
Figure 41. Apron capacity envelope for Example 2a, scatter diagram 
Apron capacity flexibility of a certain apron configuration can be expressed as the ratio 
of the capacity provided by current apron (given configuration and operational 
constraints) and theoretical capacity allowed by an absolutely flexible apron with no 
user constraints. 
For the two examples given above we have apron capacity envelopes of the constraint-
free apron depicted in Figures 42 and 43, with a dashed line, while the solid lines are 
capacity envelopes of the current configurations.  
By using the two envelopes (of the current and constraints-free apron configurations), 
flexibility of the current configuration can be expressed as the ratio of the surfaces 
under these envelopes. For the first case, apron capacity flexibility is 0.8 and in the 
second case it is 0.5. An absolutely flexible apron has a flexibility equal to 1.0. 
0
1
2
3
4
5
6
7
8
0 1 2 3 4 5 6 7 8
ot
he
r a
irl
in
es
 (
ai
rc
ra
ft
/h
)
base airline (aircraf t/h) 
5 base a/l stands + 6 other a/l stands
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
108 
 
 
Figure 42. Apron capacity envelopes for a constraint-free apron and current configuration in Example 1 
 
Figure 43. Apron capacity envelopes for a constraint-free apron and current configuration in Example 2a 
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
ot
he
r O
/D
 c
om
b.
 (a
irc
ra
ft
/h
)
domestic O/D comb. (aircrat/h)
5 dom.f l. stands + 6 ALL f l. stands
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
ot
he
r a
irl
in
es
 (
ai
rc
ra
ft
/h
)
base airline (aircraf t/h)
5 base a/l stands + 6 other a/l stands
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
109 
 
Expressed this way, apron capacity flexibility can serve as a measure of the opportunity 
of a current apron configuration to gain more capacity by reallocation of stands between 
dominant users. It indicates to what extent the available physical resources (offered by a 
constraint-free configuration) are utilized, or, in other words, to what extent capacity 
reserves have already been exhausted. The more flexible the apron configuration is (i.e. 
closer to 1.0), the more able it is to accept changes in traffic structure. However, if 
congestion appears on the apron, a higher flexibility at the same time signifies a lower 
ability to accept for additional traffic through adequate resources reallocation. This 
means that, if an absolutely flexible apron is confronted with congestion, the only 
solution would be expansion of the capacity. Less flexible aprons are more sensitive to 
changes in traffic structure, but allow more possibility to meet (temporary or long term) 
changes in traffic volume and structure with relaxation of constraints, than do highly 
flexible aprons. 
This information can be used for planning purposes, when deciding how to manage 
resources to meet changes in demand, prior to physical expansion. It provides 
information on how close to the upper limit of the capacity it is, as well as how much 
can be done to gain additional capacity through reassignment, before expansion.  
Both Figure 42 and Figure 43 illustrate passenger apron capacity flexibility, as an 
indicator of its ability to cope with changes in user’s structure changes to which it is 
exposed during day, week or season. In other words, it can serve as a measure of the 
passenger apron’s potential to meet changes in traffic structure by users, expected on the 
medium- to long-term horizon, with adequate reallocation of available resources. 
Furthermore, if the entire apron area (not only passenger aprons) is taken into 
consideration, then the potential of the apron in accommodating changes in passenger 
traffic, of an irregular and temporary character, can also be discussed. These changes 
may be due to the occurrence of certain events, either planned in advance (medium-
term, e.g. congresses, sports competitions, etc.), or unexpected (short-term, due to e.g. 
meteorological conditions). In such situations the overflow in passenger traffic can be 
accommodated by underutilized apron areas for non-passenger users, such as cargo or 
general aviation.   
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
110 
 
5.7 Possible areas of application 
“Dealing with an uncertain future, planners, managers and designers need to recognize 
the wide range of situations that may occur, examine the implications of these scenarios, 
and develop strategies that enable them to seize opportunities and protect them from 
risk” (TRB, 1987). In this case macroscopic models have their advantage, enabling fast 
analysis of various scenarios requiring less detailed input data (but still enough to 
provide good estimates). The output from macroscopic models is not meant to be 
directly used for final decision-making, but rather for selection of the best candidate 
scenarios that will be further analyzed through high-level-of-detail simulation tools.  
The approach to airport airside modeling, aimed at airside capacity assessment and 
analysis, proposed in the thesis, is suitable for analysis of small-to-medium airports with 
the capacity constraint mainly at the terminal complex. It can also apply for a quick scan 
of the reserve capacity at airports exposed to frequent and significant changes of their 
dominant airlines/other users. Practice shows that the reasons for this can be various. 
For example, after Switzerland entered the Schengen zone, Zurich Airport (Switzerland) 
had to transform the concept of its terminal-apron complex from exclusively 
international to the Schengen/non-Schengen concept. The transitional and final concepts 
are shown in Appendix 15. During the transitional (reconstruction) period remote stands 
had to be intensively used. For such purpose, new Busgates were built in order to enable 
higher utilization of the available remote stand, which are now used for overflow of 
traffic from contact stands.  
Also, numerous airports experience changes in their uses structure, imposed by the 
constant growth of low-cost market segment.  
Another example is Belgrade Airport (Serbia) which recently changed its main 
customer, when Etihad signed the contract of strategic partnership with (former) Jat 
Airways (now Air Serbia), in August 2013. With new entrant in the market and 
favorable location from the perspective of west-east connections, there is a certain 
potential for the airport to change its role in the air transport network, from O/D into a 
regional hub.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
111 
 
Some recent tendencies in airside modeling, for the purpose of supporting strategic (as 
well as tactical) airport planning, are directed toward the integration of existing 
macroscopic (and microscopic) models covering a wide range of issues, such as 
capacity, noise, emissions, cost-benefit analysis, 3rd party risk, etc. which are important 
for decision-makers to balance between. Improvement of each of the components can 
contribute to the entire decision-making process.  
Regarding the airside capacity analysis, a group of authors (Zografos and Madas, 2007) 
propose MACAD (for strategic) and TAAM, SIMMOD or RAMSPlus (for tactical 
planning). Another group of authors (Wijnen et. Al., 2008; and Kwakkel et al., 2009) 
apply the usual approach – the runway system is taken as the constraining factor and the 
FAA Airfield Capacity Model is used for airside capacity modeling. An approach 
described in the thesis offers the possibility to include apron capacity into consideration 
when needed.  
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
112 
 
6 Conclusion 
The introductory chapters of the thesis summarized important factors that affect runway 
system, apron and overall airside capacity.  
Chapter 1 discussed airports element usage with respect to different criteria: market 
segments, physical constraints and operational constraints, together with their impact on 
the capacity of the entire airside system. Additionally, Chapter 2 elaborated different 
airport categorizations with respect to traffic parameters, giving a special emphasis on 
airport types with respect to their role in the network.  
Chapter 3 summarized the definitions of airport airside capacity and factors that affect 
it. A particular attention was given to the characteristics of hub airports, such as: 
temporal concentration, ideal wave, wave-system structure and evolutionary phases as 
important factors that affect runway-apron functional relationship and consequently the 
overall airside capacity. The reminder of this chapter also discussed the issue of peak 
period: traditional approaches and some of the latest research results. 
Chapter 4 gave a broad overview of the models for airside capacity modeling. It showed 
that the most mature areas are the runway system and entire airfield capacity modeling. 
Detailed state-of-the-art in the field of apron capacity modeling followed, being the 
main research area of this thesis.  
In Chapter 5 an approach to analyzing the capacity of the existing (built) system under 
given physical and operational constraints, for given demand characteristics, is 
proposed. The approach considers linking of the (existing) runway system capacity 
model with the (extended) apron capacity model, through the runway-apron functional 
relationship (observed and defined in the thesis for different airport types: origin-
destination and hub).  
For such a purpose, modification of the existing analytical models was proposed, by 
including both constraints on aircraft classes and users, considering later different traffic 
types as well. It was followed by apron capacity representation, using newly suggested 
apron capacity envelopes. They provide information about capacity sensitivity of a 
certain apron configuration in dependence of the demand structure with respect to the 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
113 
 
dominant users. Furthermore, utilization factor, separation time and turnaround time, 
being significant variables that influence estimated capacity figures, were discussed.  
Different numerical examples were used to show sensitivity of apron and consequently 
overall airside capacity to changes in: type of traffic, the runway system performance, 
and apron operational constraints.  
Two airport types, with respect to the role of the airport in the air transport network, 
were analyzed: origin-destination airports (serving mainly point-to-point traffic, 
resulting in traffic being distributed rather “uniformly” during the day, with more or less 
pronounced peak periods), and hub airports (serving primarily airline/alliance 
coordinated flights, resulting in traffic being concentrated in waves of flights). The 
thesis also addressed the co-existence of point-to-point and coordinated flights at a 
single airport.  
It was shown that the common approach to overall airside capacity can be applied at 
origin-destination airports: the runway system and apron may be observed 
independently of each other, deriving a conclusion on overall airside capacity by 
comparing the two of them.  
On the other hand, capacities of the runway system and apron at hub airports have to be 
observed linked to each other. It was shown that the maximum number of aircraft within 
the wave depends either on apron configuration and demand structure (static apron 
capacity), or it is limited by the given level of service (defined over maximum 
acceptable connecting time). The latter is a function of the runway system performance. 
It was also shown that turnaround times for aircraft concentrated in waves depend on 
the runway system performance.  
Consequently, an approach to estimating apron capacity at hub airport was offered. In 
addition to apron configuration and demand structure, it also takes into consideration 
wave-system structure parameters and the runway system performance.  
The difference between apron capacity for origin-destination airports and hub airports 
that serves only coordinated flights was analyzed. It was shown that the same 
infrastructure provides higher throughput for point-to-point traffic (O/D airports) than 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
114 
 
for coordinated traffic (hub airport), having the same demand structure. That is a 
consequence of longer turnaround times for aircraft concentrated in waves, resulting 
from at least minimum connecting time required between all connections.    
Further on, the difference between the theoretical apron capacity hub airport can offer 
for coordinated flights (assuming continuous exchange of departures and arrivals of the 
successive wave) and utilized apron capacity under specific wave-system structure 
which base airline/alliance operates from the hub (defined by wave repeat cycle) was 
analyzed.   
The idle periods between the waves were seen as a spare capacity that can be “offered” 
to other users (if they appear) without jeopardizing the wave-system structure. Due to 
this, utilization of apron capacity at hub airports was analyzed, assuming that point-to-
point traffic is allowed to use idle stands. Two scenarios were observed: preferential use 
(when contact stands are available for point-to-point flights between consecutive waves) 
and exclusive use (when contact stands are used only by coordinated flights). Capacity 
changes for both scenarios were shown for the full range of coordinated/point-to-point 
flights shares (100-0 to 0-100 %). As expected, the results showed higher utilization 
with the preferential use scenario, than in the case when contact stands are exclusively 
used by coordinated flights. 
Further on, the conditions under which apron capacity is sensitive to the runway system 
performance were analyzed. Results showed that a lower runway system performance 
may result in a lower maximum number of aircraft per wave, and consequently a lower 
apron and overall airside capacity. However, under certain conditions, the effects of the 
runway system performance on apron capacity may be concealed, for example when 
apron static capacity is more constraining than level-of-service capacity. 
The thesis also observed reserve/latent apron capacity, being considered as a potential 
for accepting additional demand prior to physical expansion. It was suggested to express 
apron capacity flexibility relative to apron capacity of the absolutely flexible apron 
(with no operational constraints on stand usage). On one hand, apron capacity flexibility 
is used as an indicator of the ability of the apron to respond to changes in demand 
structure. At the same time, it is considered as an indicator of latent/reserve capacity 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
115 
 
that may be “activated” by reallocation of available resources i.e. relaxation of the 
constraints on stand use.  
At the end a short discussion was given on possible areas of application of the proposed 
approach in the field of airport airside capacity analysis.  
An apron capacity estimation tool was developed as a supplementary material to the 
thesis. The general user instructions on how to use it for O/D and hub airports is 
provided in Appendix 14. 
Research results of this thesis can be further developed in different directions. The main 
avenues for expansions/modification of the models offered in the thesis for airport apron 
and overall airside capacity analysis are indicated in the list of assumptions that the 
models are based on, summarized in Section 5.1. Of all the ones listed, the two can be 
considered of higher importance.  
One is introduction of stand occupancy times as random variables to obtain estimates of 
mean and standard deviation of apron capacity. Similar approach as in Bandara and 
Wirasinghe (1988) can be applied. 
The other is evaluation of the scenarios for gaining additional capacity through 
operational constraints relaxation, based on the cost of their implementation. Associated 
costs depend on the type of transformation proposed (terminal building reconstruction, 
implementation of the new technologies in passenger processes, reallocation of the 
stands between traditional airlines and LCCs, relaxation of stand use policy from 
exclusive to preferential or to partially/entirely joint use, etc.). The best scenario(s) are 
those that provides capacity/demand match at acceptable costs. 
The models offered in the thesis can also be further adapted for airside capacity analysis 
of passenger airports with specific operational requirements (e.g. de-icing process is 
done at the aircraft stand), or other types of airports with respect to demand 
characteristics (e.g. seasonal airports, airports with mixed passenger/non-passenger 
market segments, etc.).  
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
116 
 
7 List of abbreviations 
AC Advisory Circular 
ACATS Airport Capacity Analysis Through Simulation 
ACRP Airport Cooperative Research Program 
AIP Aeronautical Information Publication 
AMAN Arrival Manager 
ARRtw ARRival time window 
ATC Air Traffic Control 
ATM Air Transport Management 
BAA British Airports Authority 
CALM Computer-assisted Approach and Landing Management 
CJ Conventional Jet Aircraft 
CLOU Co-operative Local resOUrce planner 
CUMA Civil Use of Military Aerodromes 
DEPtw DEParture time window 
DFS Deutsche Flugsicherung 
DMAN Departure Manager 
EC European Commission 
EU European Union 
EUR Continental flights 
FAA Federal Aviation Administration 
HERMES Heuristic Runway Movement Event Simulation 
IATA International Air Transport Association 
ICA Intercontinental flights 
ICAO International Civil Aviation Organization 
ICT Intermediate Connecting Time 
JFK John F. Kennedy International Airport 
LCC Low-Cost Carrier 
LEONARDO Linking Existing on Ground, Arrival and Departure Operations 
LOS Level Of Service 
MACAD Mantea Airfield Capacity and Delay 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
117 
 
 MAP Military Airport Programme 
MARS Multi Aircraft Ramp System Concept 
MaxCT Maximum Connecting Time 
MCT Minimum Connecting Time 
MLS Microwave Landing System 
MUC Munich International Airport 
NLA New Large Aircraft 
OAG Official Airline Guide 
O/D Origin-Destination 
RAMS Reorganized ATC Mathematical Simulator 
SEAP Steeper Approach Procedure 
SGAP Staggered Approach Procedure 
SMAN Surface Manager 
SFO San Francisco International Airport 
SIMMOD Airport and Airspace Simulation Model 
SOT Stand Occupancy Time 
TAAM Total Airspace and Airport Modeler 
TAM Total Airport Manager 
TAT TurnAround Time 
TMAN Turnaround Manager 
TRB Transportation Research Board 
WB Wide Body aircraft 
WL Wave Length 
WRC Wave Repeat Cycle 
WNX Weighted indirect connection number 
ZRH Zurich International Airport 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
118 
 
8  List of variables and parameters 
ARRtw arrival time window - the time period during which aircraft 
(concentrated in a wave) arrive 
ARRDEPtw time window for alternating arrivals and departures 
C apron capacity – saturation apron capacity 
Ci apron capacity limited by the ith group of stands (designated for user 
i, or for aircraft class i) 
Cij apron capacity limited by the group of stands available for user i and 
aircraft class j 
CI apron capacity limited by the group of stands available for 
coordinated flights 
CII apron capacity limited by the group of stands available for point-to-
point flights 
CT theoretical apron capacity – saturation apron capacity at hub airport 
serving only coordinated flights (assumes completely overlapped 
waves) 
CU utilized apron capacity at hub airport serving only coordinated flights, 
accounting for demand requirement (defined by the wave repeat 
cycle) 
DEPtw departure time window – the time period during which aircraft 
(concentrated in a wave) depart, after being served, including 
transfers between flights 
MCT minimum connecting time – minimum time period to facilitate 
transfers between two connections 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
119 
 
MCTC minimum time period to facilitate transfers between continental 
connections 
MCTIC minimum time period to facilitate transfers involving intercontinental 
connections 
MaxCT maximum acceptable connecting time as a measure of level of service 
MaxCTC maximum acceptable connecting time between continental 
connections 
MaxCTIC maximum acceptable connecting time involving intercontinental 
connections 
iµ  utilization factor of the ith group of stands (designated for user i, or 
for aircraft class i) 
Nall total number of available stands 
Ni number of stands designed for aircraft class i (only) 
Ni’ number of stands in the i
th group of stands  (stands allowed to be used 
by user i, or stands designed for aircraft class i and for aircraft larger 
than i) 
Nij’ number of aircraft stands that may be used by aircraft of user i and 
class j (stands allowed to be used by user i, designed for aircraft class 
j and for aircraft larger than j) 
N maximum number of aircraft in a wave 
Na/c static apron capacity – maximum simultaneous number of aircraft at 
the apron, accounting for apron configuration and demand structure 
Nij maximum simultaneous number of aircraft at the apron limited by the 
group of stands available for user i and aircraft class j 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
120 
 
Nlos maximum number of aircraft that may be scheduled within a wave to 
satisfy given level of service (defined by maximum acceptable 
connecting time - MaxCT) 
NlosC maximum number of continental aircraft that may be scheduled 
within a wave to satisfy given level of service (defined by maximum 
acceptable connecting time - MaxCTC) 
NlosIC maximum number of intercontinental aircraft that may be scheduled 
within a wave to satisfy given level of service (defined by maximum 
acceptable connecting times – MaxCTC and MaxCTIC) 
ip  proportion of aircraft of user/class i in the population of aircraft 
demanding service 
ijp  proportion of aircraft of user i and class j in the population of aircraft 
demanding service 
Is  share of coordinated traffic in total demand 
'ijs  
cumulative share user/aircraft class combination allowed to use the 
ijth group of stands 
SOT average stand occupancy time as a sum of average turnaround time 
(TAT) and average separation time (ST) 
SOTC average stand occupancy time for continental flights 
SOTC average stand occupancy time for intercontinental flights 
ST separation time between two consecutive aircraft stand occupancies, 
accounting for positioning time and apron taxiway/taxilane system 
leading to/from aircraft stands 
TAT average turnaround time 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
121 
 
TATC average turnaround time for continental flights 
TATIC average turnaround time for intercontinental flights 
t  weighted average stand occupancy time of all aircraft demanding 
service 
'it  weighted average stand occupancy time of all aircraft allowed to use 
the ith group of stands 
'ijt  weighted average stand occupancy time of all user/aircraft class 
combination allowed to use the ijth group of stands 
Ti average stand occupancy time of the aircraft of user i or class i 
Tij average stand occupancy time of the aircraft of user i and class j 
At  
weighted average separation time of all landing aircraft on the 
runway 
ij
ji
ijA tpt ∑ ⋅=
,  
ijt - minimum separation time between aircraft i and j (separation time 
on approach between aircraft i and j , or runway occupancy time of 
aircraft i, whichever is greater) 
ijp - probability of (i,j) pair appearance jiij ppp ⋅=  
Dt  weighted average separation time between all departing aircraft on 
the runway 
ij
ji
ijD tpt ∑ ⋅=
,
'
 
ijt' - minimum separation time between aircraft i and j on take-off 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
122 
 
p
At  
weighted average separation time of all landing aircraft on the 
runway allowing one departure to be performed between any two 
consecutive arrivals 
∑ ⋅=
ji
p
ijijp tpt
,
 
p
ijt - minimum separation time between aircraft i and j on approach 
that allows releasing one departure between them 
U utilization of the theoretical capacity at pure hub airports with respect 
to demand requirements 
WL wave length - the total length of a wave of flights including  arrival 
time window (ARRtw), minimum connecting time (MCT) and 
departure time window (DEPtw) 
WRC wave repeat cycle – the time period between the same points of the 
consecutive waves, e.g. beginning of the arrival time window 
 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
123 
 
9 References 
 
[1] AirTOpsoft (2013) http://www.airtopsoft.com/products.html (last accessed: 
August 2013) 
[2] Airtravel about (2013) http://airtravel.about.com/od/airports/a/ 
minimumconnect.htm (last accessed: June, 2013) 
[3] Anagnostakis, I., Idris, H.R., Clerke, . J.-P., Feron, E., Hansman, R.J., Odoni, 
A.R. and Hall W.D. (2000) A Conceptual Design of A Departure Planner 
Decision Aid. Proceedings of the 3rd USA/Europe Air Traffic Management 
R&D Seminar, Napoli, Italy 
[4] Andreatta, G., Brunetta, L., Odoni, A., Righi, L., Stamatopoulos, M.A. and 
Zografos, K.G. (1999) A Set of Approximate and Compatible Models for 
Airport Strategic Planning on Airside and Landside. Air Traffic Control 
Quarterly 7 (4): 291-317 
[5] Ashford, N.J., Mumayiz, S.A. and Wright, P.H. (2011) Airport Engineering (4th 
edition). John Wiley & Sons Inc., U.S. 
[6] Ashford, N.J., Stanton H.P.M., Moore C.A. (1997) Airport Operations (2nd 
edition), McGraw-Hill, New York, U.S. 
[7] Ashford, N.J., Stanton H.P.M., Moore C.A, Coutu P., Beasley, J.R. (2013) 
Airport Operations (3rd edition). McGraw-Hill, New York, U.S. 
[8] Ashford, N.J., Wright, P.H. (1992) Airport Engineering (3rd edition). John 
Wiley & Sons Inc. Hoboken, U.S. 
[9] ATAC (2013) http://www.atac.com/simmod-pro.html (last accessed: August, 
2013) 
[10] Atkins, S. and Brinton, C. (2002) Concept Description and development Plan for 
the Surface Management System. Journal of Air Traffic Control 44 (1) 
[11] Babic, O., Teodorovic, D. and Tosic, V. (1984). Aircraft Stand Assignment to 
Minimize Walking. Journal of Transportation Engineering 110 (1): 55-66 
[12] Bandara, S. and Wirasinghe, S.C. (1988) Airport Gate Position Estimation under 
Uncertainty. Transportation Research Record 1199: 41-48 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
124 
 
 
[13] Barco (2004) CALM - Computer-assisted Approach and Landing Management 
for Zurich Airport, PDF brochure http://www.barco.com/barcoview/ 
downloads/OSYRIS_CALM.pdf (last accessed: August, 2013) 
[14] Barrer, J.N. (2007) A New Method for Modeling Complex Runway Systems, 
Proceedings of the 7th USA/Europe Air Traffic Management R&D Seminar, 
Barcelona, Spain 
[15] Barrer, J.N. and Kuzminski, P.C. (2005) Analyzing the Runway Capacity of 
Complex Airports, Proceedings of the 6th USA/Europe Air Traffic Management 
R&D Seminar, Baltimore, MD, U.S. 
[16] Beasley, J.E., Krishnamoorthy, M. Sharaiha, Y.M., Abramson, D. (2000) 
Scheduling Aircraft Landings – The Static Case. Transportation Science 34 (2): 
180-197 
[17] Blumstein, A. (1959) The Landing Capacity of a Runway. Operations Research 
7 (1): 752–763 
[18] Boehme, D. (2005). Tactical Departure Management with the Eurocontrol/DLR 
DMAN. Proceedings of the 6th USA/Europe Air Traffic Management R&D 
Seminar, Baltimore, Maryland, U.S. 
[19] Boehme, D., Brucherseifer, R. and Christoffels, L. (2007) Coordinated Arrival 
Departure Manager. Proceedings of the 7th USA/Europe Air Traffic 
Management R&D Seminar, Barcelona, Spain 
[20] Burghouwt, G. (2007) Airline Network Development in Europe and its 
Implications for Airport Planning. Ashgate, Aldershot, England 
[21] Burghouwt, G., de Wit, J. (2005) Temporal configuration of European airline 
networks. Journal of Air Transport Management 11: 185-198 
[22] Burghouwt, G. (2013) Airport Capacity Strategies in the Era of Airline Multi-
hub Networks. Discussion Paper No. 2013-5 at International Transport Forum. 
[23] Carpatair (2013) http://www.carpatair.com/Transfer_time_Minimum_ 
connecting_time/ (last accessed: June, 2013) 
[24] Castillo-Manzano, J.I., Lopez-Valpuesta, L., Pedregal, D.J. (2012) What role 
will hubs play in the LCC point-to-point connections era? The Spanish 
experience. Journal of Transport Geography 24: 262-270 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
125 
 
[25] Caves, R. (1994). A Search for More Airport Apron Capacity. Journal of Air 
Transport Management 1 (2): 109-120 
[26] Cheng Y. (1997) A Knowledge-Based Gate Assignment System Integrated with 
Mathematical Programming. Computers and Engineering 32 (4): 837-852 
[27] Cherniavsky, E.A. and Abrahamsen, T.R. (2000) Aviation System Performance 
Metrics: Airport Utilization (sponsored by FAA). The MITRE Corporation 
[28] Danesi, A. (2006) Measuring airline hub timetable coordination and 
connectivity: definition of a new index and application to a sample of European 
hubs. European Transport 34: 54-74   
[29] De Barros, A.G. and Wirasinghe, S.C. (2004) New Large Aircraft Gate 
Requirements for Shared Space and Stage Construction. Journal of 
Transportation Engineering 130 (2): 141-149 
[30] De Neufville, R. (2008) Low-Cost Airports for Low-Cost Airlines: Flexible 
Design to Manage the Risks. Transportation Planning and Technology 31:1,  
35-68 
[31] De Neufville, R., Belin, S.C. (2002) Airport Passenger Buildings: Efficiency 
through Shared Use of Facilities. Journal of Transportation Engineering 128 
(3): 201-210 
[32] De Neufville, R., Odoni, A. (2003) Airport Systems - Planning, Design and 
Management. McGraw-Hill, New York, U.S. 
[33] De Wit, J.G., Zuidberg, J. (2012) The growth limits of the low cost carrier 
model. Journal of Air Transport Management 21: 17-23 
[34] Dennis, N. (1994) Airline hub operations in Europe. Journal of Transport 
Geography 2(4): 219-233 
[35] Dennis N. (2005) The development of an airport as a hub. Lecture notes, 
academic year 2005/06, University of Westminster, London, U.K. 
[36] Deutsche Bank Research (2006) The future of the hub strategy in the air 
transport industry. Commissioned by Unique (Zurich Airport)  
http://www.dbresearch.de/PROD/DBR_INTERNET_DE-PROD/PROD 
0000000000201837.pdf (last accessed: August, 2013) 
[37] DFS -  Deutsche Flugsicherung (2004) Aeronautical Information Publication 
(AIP), Germany 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
126 
 
 
[38] Ding, H., Lim, A., Rodrigues, B. and Zhu, Y. (2005) The Over-constrained 
Airport Gate Assignment Problem. Computers & Operations Research 32: 
1867-1880 
[39] Dobruszkes, F. (2006) An analysis of European low-cost airlines and their 
networks. Journal of Transport Geography 14: 249-264 
[40] Dobruszkes, F. (2009) New Europe, new low-cost services. Journal of 
Transport Geography 17: 423-432 
[41] Dorndorf, V., Drexl, A., Nikolin, Y. and Pesch, E. (2007) Flight Gate 
Scheduling: State-of-the-Art and Recent Developments. Omega 35: 326-334 
[42] EUROCONTROL (2007) A Place to Stand: Airports in the European Air 
Network. EUROCONTROL Trends in Air Traffic, Volume 3 
[43] European Commission (2005) Official Journal of the European Union. Vol. 48 
http://eur-lex.europa.eu/LexUriServ/site/en/oj/2005/c_312/c_31220051209en 
00010014.pdf (last accessed: August, 2013) 
[44] European Commission DG-TREN (2006) SPADE - Supporting Platform for 
Airport Decision-Making and Efficiency Analysis, publishable Final Activity 
Report  
[45] European Commission DG-TREN (2008) ASSET – Aeronautic Study on 
Seamless Transport, Report on State of the Art, DLR and ASSET Partners 
[46] FAA – Federal Aviation Administration (2013) http://www.faa.gov/ 
airports/planning_capacity/passenger_allcargo_stats/categories/ (last accessed: 
August, 2013) 
[47] FAA – Federal Aviation Administration (1983) Airport Capacity and Delay. 
Advisory Circular, AC 150/5060-5 
[48] Feron, E., Hansman, R.J., Odoni, A.R., Cots, R.B., Delcaire, B., Feng, X., Hall, 
W.D., Idris, H.R., Muharremoglu, A. and Pujet, N. (1997). The Departure 
Planner: A Conceptual Discussion. Massachusetts Institute of Technology, 
Cambridge, U.S. 
[49] Forsyth, P. (2007) The impacts of emerging aviation trends on airport 
infrastructure. Journal of Air Transport Management 13: 45-52 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
127 
 
 
[50] Franke, M. (2004) Competition between network carriers and low-cost 
carriers—retreat battle or breakthrough to a new level of efficiency? Journal of 
Air Transport Management 10: 15-21 
[51] Gass, S.I. (1992) Evaluation of Air Traffic Modeling Tools: Validation and 
Review of Results and Documentation (submitted to Federal Aviation 
Administration) 
[52] Genc, H.M., Erol, O.K., Eksin, I., Berber, M.F. (2012) A stochastic 
neighborhood search approach for airport gate assignment problem. Expert 
Systems with Applications 39: 316-327 
[53] Gilbo, E.P. (1993). Airport Capacity: Representation, Estimation, Optimization. 
IEEE Transactions on Control Systems Technology 1( 3): 144-154 
[54] Guenther, Y., Inard, A., Werther, B., Bonnier, M., Spies, G., Mersden, A., 
Temme, M., Boehme, D., Lane, R. and Niederstrasser, H. (2006). Total Airport 
Management - Operational Concept and Logical Architecture 1.0. 
DLR/Eurocontrol EEC, Braunschweig, Germany/ Paris, France 
[55] Haghani, A. and Chen, M.C. (1998) Optimizing Gate Assignments at Airport 
Terminals. Transportation Research Part A 32 (6): 437-454 
[56] Halpern, N. (2011) Measuring seasonal demand for Spanish airports: 
Implications for counter-seasonal strategies. Research in Transportation 
Business and Management 1: 47-54 
[57] Hamzawi, S.G. (1986) Management and Planning of Airport Gate Capacity: A 
Microcomputer-Based Assignment Model. Transportation Planning and 
Technology 11: 189-202 
[58] Harris, R.M. (1972) Models for Runway Capacity Analysis. Report No. FAA-
EM-73-5 (prepared for Department of Transportation, Federal Aviation 
Administration). The Mitre Corporation, Virginia, U.S.  
[59] Hassounah, M. and Steuart, G.N. (1993) Demand for Aircraft Gates. 
Transportation Research Record 1423: 26-33  
[60] Hesselink, H. and Basjes. N. (1998) Mantea departure Sequencer: Increasing 
Airport Capacity by Planning Optimal Sequences. Proceedings of the 2nd 
USA/Europe Air Traffic Management R&D Seminar, Orlando, Florida, U.S.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
128 
 
[61] Hockaday, S.L.M. and Kanafani, A.K. (1974) Developments in Airport Capacity 
Analysis. Transportation Research 8: 171-180. 
[62] Horonjeff, R. (1975) Planning and Design of Airport (2nd edition). McGraw-
Hill, New York, U.S. 
[63] Horonjeff, R. and F.X. McKelvey. (1994). Planning and Design of Airports (4th 
edition). McGraw-Hill, New York, U.S. 
[64] Horonjeff, R., McKelvey, F.X.. Sproul, W.Y. and Young. S.P. (2010) Planning 
and Design of Airports (5th edition). McGraw-Hill, New York, U.S. 
[65] IATA - International Air Transport Association (2004) Airport Development 
Reference Manual (9th edition). International Air Transportation Association, 
Montreal, Canada 
[66] IATA - International Air Transport Association (2008) Total AirportSim 
brochure, http://www.iata.org/ps/publications/pages/totalairportsim.aspx (last 
accessed: October, 2008) 
[67] Inform (2013) http://www.inform-software.com/products/groundstar/core-
processes/gs-hubcontrol/ (last accessed: Augut, 2013) 
[68] ISA Software (2013) http://www.ramsplus.com/ (last accessed: August, 2013) 
[69] Janic, M. (2004) Expansion of Airport Capacity: Case of London Heathrow 
Airport. Transportation Research Record 1888: 7-14 
[70] Janic, M. (2008) Modelling the Capacity of Closely-Spaced Parallel Runways 
Using Innovative Approach Procedures. Transportation Research Part C 16 (6): 
704-730 
[71] Janic, M. (2009) Airport Analysis, Planning and Design: Demand, Capacity and 
Congestion. Nova Science Publishers, Inc. New York, U.S. 
[72] Jeppesen (2013) http://www.jeppesen.com/industry-solutions/aviation/ 
government/total-airspace-airport-modeler.jsp (last accessed: August, 2013) 
[73] JFK - John F. Kennedy International Airport , The Port Authority of New York 
and New Jersey (2012) http://www.panynj.gov/airports/jfk-airport-map.html 
(last accessed: February 2012) 
[74] Jonge, H., Tuinstra, E. and Seljee, R. (2005) Outbound Punctuality Sequencing 
by Collaborative Departure Planning. Proceedings of the 6th USA/Europe Air 
Traffic Management R&D Seminar, Baltimore, Maryland, U.S. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
129 
 
 
[75] Kraus, A., Koch, B. (2006) The opportunities and threats of turning airports into 
hubs. Journal of Air Transportation 11 (1): 87-112 
[76] Kwakkel, J.H., Wijnen, R.A.A., Walker, W.E. and Marchau, V.A.W.J. (2009) A 
Fast and Simple Model for a Quick Scan of Airport Performance. Proceedings of 
the 13th Air Transport Research Society Conference, Abu Dhabi, United Arab 
Emirates 
[77] Lawson, D. (1997) Surface Movement Advisor (SMA). Proceedings of the 1st 
USA/Europe Air Traffic Management R&D Seminar, Saclay, France  
[78] Malighetti, P., Paleari, S., Redondi, R. (2008) Connectivity of the European 
airport network: ‘‘Self-help hubbing’’ and business implications. Journal of Air 
Transport Management 14: 53-65 
[79] Malighetti, P., Paleari, S., Redondi, R. (2009) Airport classification and 
functionality within the European network. Problems and Perspectives in 
Management 7 (1): 183-196 
[80] Mangoubi, R.S. and Mathaisel, D.F.X. (1985) Optimizing Gate Assignment at 
Airport Terminals. Transportation Science 19 (2): 173-188 
[81] Mirkovic, B. (2010) Airfield modelling – state of the art. Proceedings of the 14th 
Air Transport Research Society Conference, Porto, Portugal 
[82] Mirkovic, B. (2011a) Analysis of airport element usage with respect to different 
users. Proceedings of 38th SYM-OP-IS-a (Symposium of Operations Research), 
pp. 697-700 (session: Traffic, transport and communication), Zlatobor, Srbija, 4-
7. October, 2011 (in Serbian) 
[83] Mirkovic, B. (2011b). Airport Apron Capacity Estimation – Model 
Enhancement. Proceedings of 14th EWGT (EURO Working Group on 
Transportation) Meeting, 26th MEC (Mini-EURO) Conference and 1st 
European Scientific Conference on Air Transport, Poznan, Poland, September 6-
9, 2011 
[84] Mirkovic, B. and Tosic, V. (2012) Airports II. University of Belgrade-Faculty of 
Transport and Traffic Engineering (textbook in Serbian) 
 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
130 
 
 
[85] Mirkovic, B., Tosic, V. (2013). Airport Apron Capacity – Estimation, 
Representation and Flexibility. Journal of Advanced Transportation special 
issue: “Methodological Advances in European Transport Research” (DOI: 
10.1002/atr.1250) 
[86] Mirkovic, B., Vidosavljevic, A. and Tosic, V. (2013) A tool to support resource 
allocation at small-to-medium seasonal airports. Proceedings of the 17th Air 
Transport Research Society World Conference, Bergamo, Italy 
[87] MUC - Munich Airport (2013a) http://www.munich-airport.de/en/consumer 
/aufenthalt_trans/airportstop/minconntime/index.jsp (last accessed: August, 
2013) 
[88] MUC - Munich Airport (2013b) http://www.munich-airport.de/en/consumer 
/anab/auto/anfahrt/anfahrt_T2/index.jsp (last accessed: August, 2013) 
[89] Newell, G.F. (1979) Airport Capacity and Delays. Transportation Science 13(3): 
201-241 
[90] Odoni, A.R. (1991) Transportation Modeling Needs: Airports and Airspace, 
Massachusetts Institute of Technology, Cambridge, U.S. 
[91] Odoni, A.R., Bowman, J., Delahaye, D., Deyst, J., Feron, E., Hansman, R.J., 
Khan, K., Kuchar, J.K., Pujet, N., Simpson, R.W. (1997) Existing and Required 
Modeling Capabilities for Evaluating ATM Systems and Concepts, 
Massachusetts Institute of Technology, Cambridge, U.S. 
[92] Odoni, A.R., Simpson, R.W. (1979) Review and Evaluation of National 
Airspace System Models, Massachusetts Institute of Technology, Cambridge, 
U.S. (prepared for U.S. Department of Transportation, Federal Aviation 
Administration) 
[93] Oettl, G., Boeck, P. (2012) An approach to airport categorization – category 
definition and quantification for air traffic related simulation purposes. Journal 
of Aerospace Operations 1: 155-170 
[94] Oettl, G., Reeb, F. (2012) Traffic related representative airport categories. 
Proceedings of Air Transport Research Society (ATRS) Conference, Tainan, 
Taiwan 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
131 
 
 
[95] Office of the Federal Register (U.S.), 49 U.S. Code, Section 47102, Definition 
(2011) http://www.gpo.gov/fdsys/search/pagedetails.action?browsePath=Title+ 
49%2FSubtitle+Vii%2FPart+B%2FChapter+471%2FSubchapter+I%2FSec.+47
102&granuleId=USCODE-2011-title49-subtitleVII-partB-chap471-subchapI-
sec47102&packageId=USCODE-2011-title49&collapse=true&fromBrowse 
=true (last accessed: August, 2013) 
[96] Pick, A. (2007). Validation Results of Airport Total Operations Planner 
Prototype CLOU. Proceedings of the 7th USA/Europe Air Traffic Management 
R&D Seminar, Barcelona, Spain 
[97] Pina, P., de Pablo, J.M. and Mas, M. (2005) Linking Existing On Ground, 
Arrival and Departure Operations, Proceedings of the 6th USA/Europe Air 
Traffic Management R&D Seminar, Baltimore, Maryland, U.S. 
[98] Radovanovic, D. (2009) Low-cost terminals. Undergraduate thesis, University of 
Belgrade-Faculty of Transport and Traffic Engineering (in Serbian) 
[99] Redondi, R., Malighetti, P. and Paleari, S. (2010) De-hubbing cases and 
recovery patterns. Proceedings of the 14th World Conference of Air Transport 
Research Society (ATRS), Porto, Portugal 
[100] Redondi, R., Malighetti, P., Paleari, S. (2012) De-hubbing of airports and their 
recovery patterns. Journal of Air Transport Management 18: 1-4 
[101] Saleh, J. H., Mark, G. and Jordan, N. C. (2009) Flexibility: a multi-disciplinary 
literature review and a research agenda for designing flexible engineering 
systems. Journal of Engineering Design 20 (3): 307-323 
[102] SFO - San Francisco International Airport (2012).  http://www.flysfo.com/ 
web/page/atsfo/airport-maps/ (last accessed: February 2012) 
[103] Siluri Integration Ltd. on behalf of DCMAC, EUROCONTROL (2008) A study 
on Existing legislation, Regulation, Procedures and Practices across ECAC and 
Relavant Non-ECAC States in the Joint Civil-Military use of Military 
Aerodromes 
[104] Steuart, G.N. (1974) Gate Position Requirements at Metropolitan Airports. 
Transportation Science 8 (2): 169-190  
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
132 
 
 
[105] Stomatopulos, M.A., Zografos, K.G. and Odoni, A. (2004) A Decision Support 
System for Airport Strategic Planning. Transportation Research Part C 12: 91-
117 
[106] Swedish, W. (1981) Upgraded FAA Airfield Capacity Model. Vols. 1 and 2. The 
MITRE Corporation, McLean, Virginia, U.S. 
[107] Tosic, V. and Horonjeff, R. (1976) Effect of Multiple Path Approach Procedures 
on Runway Landing Capacity. Transportation Research 10: 319-329 
[108] TRB - Transportation Research Board (2012a) Airport Cooperative Research 
Program, Report 79: Preparing Peak Period and Operational Profiles - 
Guidebook. Washington, U.S. 
[109] TRB - Transportation Research Board (2012b) Airport Cooperative Research 
Program, Report 82: Evaluating Airfield Capacity. Washington, U.S. 
[110] TRB - Transportation Research Board, National Research Council (1987) 
Measuring Airport Landside Capacity. Special Report No. 215. Washington, 
U.S. 
[111] Venkatakrishnan, C.S., Barnett, A., Odoni, A.R. (1993) Landings at Logan 
Airport: Describing and Increasing Airport Capacity. Transportation Science 17 
(3): 211-227 
[112] Vienna Airport (2013) http://www.viennaairport.com/jart/prj3/va/main. 
jart?rel=en&content-id=1249344074235&reserve-mode=active (last accessed: 
June, 2013) 
[113] Wijnen, R.A.A., Warren, E.W. and Kwakkel, J.H. (2008) Decision Support for 
Airport Strategic Planning. Transportation Planning and Technology 31(1): pp. 
11-34 
[114] Wirasinghe, S.C. and Bandara, S. (1990) Airport Gate Position Estimation for 
Minimum Total Costs – Approximate Closed Form Solution. Transportation 
Research B 24B(4): 287-297  
[115] Zografos, K.G. and Madas, M. A. (2007) Advanced Modelling Capabilities for 
Airport Performance Assessment and Capacity Management. Proceedings of the 
86th TRB Annual Meeting, Washington D.C., U.S. 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
133 
 
 
[116] ZRH - Zurich Airport International (2008) Zurich 2010: Schengen -
Compatibility and Centralization of Security Checkpoints (presentation) 
[117] ZRH - Zurich Airport International (2012a) http://www.zurich-
airport.com/desktopdefault.aspx/tabid-79/ (last accessed: February, 2012) 
[118] ZRH - Zurich Airport International (2012b) Airside User Concept Winter 
Timetable 2011 (presentation) 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
134 
 
10 Bibliography 
 
[1] Alderighi, M., Cento, A., Nijkamp, P., Rietveld, P. (2005) Network 
competition—the coexistence of hub-and-spoke and point-to-point systems. 
Journal of Air Transport Management 11: 328-334 
[2] Anagnostakis, I., Clarke, J-P., Boehme, D., Voelckers, U. (2001) Runway 
Operations Planning and Control Sequencing and Scheduling. Proceedings of 
the 34th Annual Hawaii Conference on System Sciences, IEEE Computer 
Society, January 3-6, 2001, Maui, Hawaii 
[3] Andersson, K., Carr, F., Feron, E. and Hall, W.D. (2000) Analysis and Modeling 
of Ground Operations at Hub Airports, Proceedings of the 3rd USA/Europe Air 
Traffic Management R&D Seminar, Napoli, Italy 
[4] Andreatta, G., Brunetta, L., Odoni, A., Righi, L., Stamatopoulos, M.A. and 
Zografos, K.G. (1998) A Set of Approximate and Compatible Models for 
Airport Strategic Planning on Airside and Landside. Proceedings of the 2nd 
USA/Europe Air Traffic Management R&D Seminar, Orlando, FL, U.S. 
[5] Artiouchine, K., Baptiste, P., Duerr, C. (2008) Runway sequencing with holding 
patterns. European Journal of Operations Research 189: 1254-1266 
[6] Atkins, S., Capozzi, B., Churchill, A., Fernandes, A., Provan, C. (2013) 
Investigation of NASA’s Spot and Runway Departure Advisor Concept at PHL, 
CLT, and LAX Airports. Proceedings of the 10th USA/Europe Air Traffic 
Management R&D Seminar, Chicago, Illinois, U.S. 
[7] Badanik, B., Laplace, I., Lenoir, N., Malavolti, E., Tomova, A., Kazda, A. 
(2010) Future Strategies for Airports. Proceedings of 27th International Congress 
of the Aeronautical Sciences, Nice, France 
[8] Barnhart, C., Belobaba, P. and Odoni, A.R., 2003. Applications of Operations 
Research in the Air Transport Industry. Transportation Science 37 (4): 368–391 
[9] Barrett, S.D. (2004) How do the demands for airport services differ between 
full-service carriers and low-cost carriers? Journal of Air Transport 
Management 10: 33-39 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
135 
 
[10] Berster, P., Gelhausen, M., Wilken, D. (2010) New Empirical Evidence on 
Airport Capacity Utilization: Relationships between Hourly and Annual Air 
Traffic Volumes. Proceedings of the 14th World Conference of Air Transport 
Research Society (ATRS), Porto, Portugal 
[11] Betancor, O., Jimenez, J.L., Socorro, M.P. (2009) The costs of peak-load 
demand: An application to Gran Canaria airport. Journal of Air Transport 
Management 16(2): 59-64 
[12] Boeck, P., Hornung, M. (2012) Airport Capacity Representation Based on 
Airside Simulations. Proceedings of 28th International Congress of the 
Aeronautical Sciences, Brisbane, Australia 
[13] Bolat, A. (2000) Procedures for Providing Robust Gate Assignments for 
Arriving Aircrafts. European Journal of Operational Research 120: 68-80 
[14] Burghouwt, G. (2013) Airport Capacity Strategies in the Era of Airline Multi-
hub Networks. Discussion Paper No. 2013-5 at International Transport Forum 
[15] Button, K. (2002) Debunking some common myths about airport hubs. Journal 
of Air Transport Management 14: 177-188 
[16] Capri, S., Ignaccolo, M. Genetic algorithms for solving the aircraft-sequencing 
problem: the introduction of departures into the dynamic model. Journal of Air 
Transport Management 10: 345-351 
[17] Caves, R.E., Gosling, G.D. (1999) Strategic Airport Planning. Pergamon, 
Elsevier Science, Amsterdam, The Netherlands 
[18] De Neufville, R., Scholtes, S. (2011) Flexibility in Engineering Design. 
Massachusetts Institute of Technology 
[19] Deau, R., Gotteland, J.B. and Durand, N. (2009) Airport Surface Management 
and Runways Scheduling, Proceedings of the 8th USA/Europe Air Traffic 
Management R&D Seminar, Napa, California, U.S.  
[20] Dennis, N. (1998) Competition for connecting traffic between hub airports in 
Europe. Proceedings of the 8th World Conference on Transport Research, 
Antwerp, Belgium  
[21] Desart, B., Gillingwater, D. and Janic, M. (2008) The Airport Capacity/Stability 
Paradox: A Feasibility Analysis of its Correlation with Delay. Proceedings of 
the 12th Air Transport Research Society Conference, Athens, Greece 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
136 
 
 
[22] Divaris, C. and Psaraki-Kalouptsidi, V. (2008) Estimation of the Runway 
Capacity of the Athens International Airport Using Analytical and Simulation 
Models. Proceedings of the 12th Air Transport Research Society Conference, 
Athens, Greece 
[23] Doble, N.A., Timmerman, J. Carniol, T., Tanino, M., Sud, V. (2009) Linking 
Traffic Management to the Airport Surface. Proceedings of the 8th USA/Europe 
Air Traffic Management R&D Seminar, Napa, California, U.S. 
[24] Edwards, J.W., Newell, G.F. (1969) A Study of Gate Use at Honolulu 
International Airport. Transportation Science 3 (3): 183-191 
[25] EUROCONTROL (2004) Enhancing Airside Capacity 
[26] EUROCONTROL-STATFOR (2008) Long-Term Forecast: IFR Flight 
Movements 2008-2030 
[27] European Commission DG-TREN (2003) THENA – Thematic Network on 
Airport Activities, Final Synthesis Report, Aena and THENA Partners 
[28] Evans, A. and Idris, H. (2005) Estimation of Arrival Capacity and Utilization at 
Major Airports. Proceedings of the 6th USA/Europe Air Traffic Management 
R&D Seminar, Baltimore, MD, U.S. 
[29] Fawcett, S.E., Fawcett, S.A. (1988) Congestion at Capacity-Constrained 
Airports: A Question of Economics and Realism. Transportation Journal 27 (4): 
42-54 
[30] Francis, G., Fidato, A., Humphreys, I. (2003) Airport–airline interaction: the 
impact of low-cost carriers on two European airports. Journal of Air Transport 
Management 9: 267-273 
[31] Francis, G., Humphreys, I., Fry, J. (2002) The benchmarking of airport 
performance. Journal of Air Transport Management 8: 239-247 
[32] Gillen, D., Lall, A. (2004) Competitive advantage of low-cost carriers: some 
implications for airports. Journal of Air Transport Management 10: 41-50 
[33] Gluchshenko, O. (2012) Optimization of Runway Capacity Utilization in the 
Case of General Pareto Curve. Proceedings of 28th International Congress of the 
Aeronautical Sciences, Brisbane, Australia 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
137 
 
 
[34] Gotteland, J.B., Durand, N., Alliot, J.M. and Page, E. (2001) Aircraft Ground 
Traffic Optimization. Proceedings of the 4th USA/Europe Air Traffic 
Management R&D Seminar, Santa Fe, NM, U.S. 
[35] Gu, Y. Chung, C.A. (1999) Genetic Algorithm Approach to Aircraft Gate 
Reassignment Problem. Journal of Transportation Engineering 125 (5): 384-389  
[36] Guenther, T., Hildebrandt, M., Fricke, H., Strasser, M. (2012) Contributions of 
advanced taxi time calculation to airport operations efficiency. Journal of 
Aerospace Operations 1: 95-106 
[37] Harris, R.M., (1969) Models for Runway Capacity Analysis. MITRE technical 
report MTR.4102. The MITRE Corporation, Virginia, U.S. 
[38] Huettig, G., Busch, W., Gronak, N. (1994) Growing demand and capacity of 
airports. Transportation Research Part A 28A (6): 501-509 
[39] Idris, H.R., Declaire, B., Anagnostakis, I., Hall, W.D., Clarke, J.P., Hansman, 
R.J., Feron, E. and Odoni, A.R. (1998) Observations of Departure Processes at 
Logan Airport to Support the Development of Departure Planning Tools, 
Proceedings of the 2nd USA/Europe Air Traffic Management R&D Seminar, 
Orlando, Florida, U.S.  
[40] Ignaccolo, M. (2003) A Simulation Model for Airport Capacity and Delay 
Analysis. Transportation Planning and Technology 26 (2): pp.135-170 
[41] Jacquillat, A., Odoni, A.R. (2012) A Study of Airport Congestion at JFK and 
EWR. Proceedings of the 5th International Conference on Research in Air 
Transportation, Berkeley, California, U.S. 
[42] Jo, G-S., Jung J-J., Yang, C-Y. (1997) Expert System for Scheduling in an 
Airline Gate Allocation. Expert Systems with Applications 13 (4): 275-282 
[43] Jung, Y., Hoang, T., Montoya, J., Gupta, G., Malik, W., Tobias, L. (2011) 
Performance Evaluation of a Surface Traffic Management Tool for Dallas/Fort 
Worth International Airport. Proceedings of the 9th USA/Europe Air Traffic 
Management R&D Seminar, Berlin, Germany 
[44] Kazda, A., Caves, R.E. (2007) Airport Design and Operations. Elsevier, Oxford, 
U.K. 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
138 
 
 
[45] Kim, A. and Hansen, M. (2009) Validation of Runway Capacity Models, 
Proceedings of the 8th USA/Europe Air Traffic Management R&D Seminar, 
Napa, California, U.S. 
[46] Kim, S.H., Feron, E., Clarke, J-P., Marzouli, A., Delahaye, D. (2013) Airport 
Gate Scheduling for Passengers, Aircraft, and Operations. Proceedings of the 
10th USA/Europe Air Traffic Management R&D Seminar, Chicago, Illinois 
[47] Koesters, D. (2008) Study on the usage of declared capacity at major German 
airports. RWTH Aachen University, Department of Airport and Air 
Transportation Research, Institute of Transport Science in cooperation with the 
Performance Review Unit, Eurocontrol 
[48] Kwakkel, J.H., Walker, W.E., Marchau, V.A.W.J. (2010) Adaptive Airport 
Strategic Planning. European Journal of  Transport and Infrastructure Research 
10 (3): 249-273 
[49] Lesire, C. (2010) Iterative Planning of Airport Ground Movements. Proceedings 
of the 4th International Conference on Research in Air Transportation, Budapest, 
Hungary 
[50] Lin, L.C., Hong, C.H. (2006) Operational performance evaluation of 
international major airports: An application of data envelopment analysis. 
Journal of Air Transport Management 12: 342-351 
[51] Lindh, A., Andersson, T., Vaerbrand, P. and Yuan, D. (2007) Airport Logistics – 
Integration of ATM and Ground Processes. Proceedings of the 7th USA/Europe 
Air Traffic Management R&D Seminar, Barcelona, Spain 
[52] Lohr, G.W., Brown, S., Stough, P. Eisenhawer, S. (2011) System Oriented 
Runway Management: A Research Update. Proceedings of the 9th USA/Europe 
Air Traffic Management R&D Seminar, Berlin, Germany 
[53] Madas, M.A. and Zografos, K.G. (2008) Airport Capacity vs. Demand: 
Mismatch or mismanagement? Transportation Research Part A 42: 203-226 
[54] Martinez, J.C, Trani, A.A. and Ioannou, P.G. (2001) Modelling Airside Airport 
Operations Using General-Purpose, Activity Based, Discrete-Event Simulation 
Tools. Transportation Research Record 1744: paper no. 01-3476 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
139 
 
 
[55] Martini, G., Manello, A., Scotti, D. (2013) The influence of fleet mix, ownership 
and LCCs on airports’ technical/ environmental efficiency. Transportation 
Research Part E 50: 37-52 
[56] Moser, R.F. and Correia, A.R. (2008) Twenty Years of Airport Simulation in 
Brazil: Lessons from the Past and Plans for the Future. Proceedings of the 12th 
Air Transport Research Society Conference, Athens, Greece 
[57] Murel, M., O'Connell, J.F. (2011) Potential for Abu Dhabi, Doha and Dubai 
Airports to reach their traffic objectives. Research in Transportation Business 
and Management 1: 36-46 
[58] Oberheid, H., Soeffker, D. (2008) Cooperative Arrival Management in Air 
Traffic Control - A Coloured Petri Net Model of Sequence Planning. 
Proceedings of the 29th International Conference Petri Nets 2008, Xi’an, China, 
June 23-27, 2008 
[59] Odoni, A., Morisset, T., Drotleff, W., Zock, A. (2011) Benchmarking Airport 
Airside Performance FRA vs. EWR. Proceedings of the 9th USA/Europe Air 
Traffic Management R&D Seminar, Berlin, Germany 
[60] Oreschko, B. Kunze, T., Schultz, M., Fricke, H. (2012) Turnaround Prediction 
with Stochastic Process Times and Airport Specific Delay Pattern. Proceedings 
of the 5th International Conference on Research in Air Transportation, Berkeley, 
California, U.S. 
[61] Pels, E., Nijkamp, P., Rietveld, P. (2001) Relative Efficiency of European 
Airports.  Transport Policy 8: 183-192 
[62] Poltoracki, I. (2006) Functional Relationship of Elements of Apron and 
Terminal Building at the Airport. Proceedings of the 2nd ICRAT, Belgrade, 
Serbia 
[63] Rakas, J., Schonfeld, P. (2004) Deterministic Models for Degraded Airside: 
Capacity and Delays. Journal of Transportation Engineering 130 (5): 545-554 
[64] Reynolds-Feighan, A., Button, K.J. (1999) An assessment of the capacity and 
congestion levels at European airports. Journal of Air Transport Management 5: 
113-134 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
140 
 
 
[65] Rietveld, P., Martijn, B. (2001) Quality of hub-and-spoke networks; the effects 
of timetable co-ordination on waiting time and rescheduling time. Journal of Air 
Transport Management 7: 241-249 
[66] Rolim, H. and Vital, J. (2009) Brazilian Methodology to Calculate Airport 
Operational Runway Capacity. Proceedings of the 13th Air Transport Research 
Society Conference, Abu Dhabi, United Arab Emirates  
[67] Sherry, L. (2009) Airport Gate and Ramp Capacity. Written exercise, SYST 
460/560, George Mason University, Center for Air Transport Systems Research 
[68] Srisaeng, P. (2010) The Responses of Traditional Airlines to Low Cost Airlines. 
Proceedings of the 4th International Conference on Research in Air 
Transportation, Budapest, Hungary 
[69] Tavakkoli-Moghaddam, R., Yaghoubi-Panah, M., Radmehr, F. (2012) 
Scheduling the sequence of aircraft landings for a single runway using a fuzzy 
programming approach. Journal of Air Transport Management 25: 15-18 
[70] Theiss, S. (2008) Support of the Determination of the Declared Capacity by the 
Use of Airside Simulation for Runway Capacity Analysis. Airlines Magazine 41 
[71] Todesco, F. and Mueller, C. (2008) Gate Assignment and Reassignment at Sao 
Paulo/Guarulhos International Airport. Proceedings of the 12th Air Transport 
Research Society Conference, Athens, Greece 
[72] TRB - Transportation Research Board (2001) Transportation Research E-
Circular No. E-C035 Airport-Airspace Simulations for Capacity Evaluation, 
Washington, U.S. 
[73] TRB - Transportation Research Board (2002) Transportation Research E-
Circular No. E-C042 Airport-Airspace Simulations: A New Outlook, 
Washington, U.S. 
[74] TRB - Transportation Research Board (2012) Airport Cooperative Research 
Program, Report 76: Addressing Uncertainty about Future Airport Activity 
Levels in Airport Decision Making. Washington, U.S. 
[75] University of Belgrade - Faculty of Traffic and Transport Engineering (2008) 
TEN CONNECT: Air Transport Bottleneck Analysis - Internal report. Belgrade, 
Serbia 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
141 
 
 
[76] Vinas Tio, M. (2010) Study of Airport Capacity vs. Efficiency, SESAR 
Challenges – Annexes, Indra Airports and Advanced Logistics Group 
[77] Vormer, F., Marsden, A., Duong, V. (2010) A Complex System Approach to 
Assessment of Airport Flexibility, Predictability and Efficiency. Proceedings of 
27th International Congress of the Aeronautical Sciences, Nice, France 
[78] Voss, W.R. and Hoffman, J. (2000) Analytical Identification of Airport and 
Airspace Capacity Constraints. Proceedings of the 3rd USA/Europe Air Traffic 
Management R&D Seminar, Napoli, Italy 
[79] Wells, A.T., Young, S.B. (2004) Airport Planning and Management (5th 
edition). McGraw-Hill, New York, U.S. 
[80] Wojcik, I.A., Mondoloni, S.L. Agbolosu-Amison, S.J., Wang, P.T.R. (2013) 
Flexibility Metrics and Their Application to Departure Queue Management. 
Proceedings of the 10th USA/Europe Air Traffic Management R&D Seminar, 
Chicago, Illinois 
[81] Wu, C.L. and Caves, R.E. (2000) Aircraft operational costs and turnaround 
efficiency at airports. Journal of Air Transport Management 6: 201-208 
[82] Wu, C.L. and Caves, R.E. (2004) Modelling and Simulation of Aircraft 
Turnaround Operations at Airports. Transportation Planning and Technology 
27: 25-46 
[83] Wu, C.L. and Caves, R.E. (2004) Modelling and Optimization of Aircraft 
Turnaround Time at an Airport. Transportation Planning and Technology 27: 
47-66 
[84] Yan, S. and Huo, C.M. (2001) Optimization of Multiple Objective Gate 
Assignment. Transportation Research Part A 3: 413-432 
[85] Yan, S. and Tang, C.H. (2007) A Heuristic Approach for Airport Gate 
Assignment for Stochastic Flight Delays. European Journal of Operational 
Research 180: 547-567 
[86] Yan, S. Chang C-M. (1998) A Network Model for Gate Assignment. Journal of 
Advanced Transportation 32 (2): 176-189 
[87] Yan, S., Chen C-Y., Tang, C-H. (2009) Airport gate reassignment following 
temporary airport closures. Transportmetrica 5 (1): 25-41 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
142 
 
 
[88] Yan, S., Shieh, C.Y. and Chen, M. (2002) A Simulation Framework for 
Evaluating Airport Gate Assignments. Transportation Research Part A 36: 885-
898 
[89] Zammit-Mangion, D., Rydell, S.A.H., Sabatini R., Jia, H. (2012) A Case Study 
of Arrival and Departure Managers Cooperation for Reducing Airborne Holding 
Times at Destination Airports. Proceedings of 28th International Congress of the 
Aeronautical Sciences, Brisbane, Australia 
[90] Zhang, Y., Chauhan, A., Chen, X. (2010) Modeling and Predicting Taxi out 
Times. Proceedings of the 4th International Conference on Research in Air 
Transportation, Budapest, Hungary 
[91] Zorgafos, K.G., Salouras, Y., Madas, M.A. (2012) Dealing with the efficient 
allocation of scarce resources at congested airports. Transportation Research 
Part C 21: 244-256 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
143 
 
11 Appendices 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
144 
 
Appendix 1 – European air transport networks 
European hub-and-spoke networks 
Hubs airports may be classified according to size of the origin-destination market, the 
stage-length of the indirect connections offered, and the geographical specialization of 
the hubs.  
Based on the size (measured through its indirect connectivity45), the main European hub 
airports, as from the traffic data from 2003, are: Frankfurt, Paris CDG, London 
Heathrow and Amsterdam (Burghouwt, 2007). “Together the four hubs accounted for 
57 per cent all indirect connections generated by the airline stations with more than 10 
indirect connections per day”. They were followed by: Madrid, Munich, Zurich, 
Copenhagen, Vienna, Rome Fuimicino, Barcelona and Milan Malpensa. Ten years later 
the situation is quite similar, only the ranking may have changed (which also depends 
on the specific measure of hubbing performance46) and some airports have lost their 
hubbing role after the bankruptcy of the base airline or its decision to withdraw (e.g. 
Milan Malpensa and Barcelona). This is a proof that airport hubbing role is highly 
dependent on airline strategy.  
Redondi et al. (2010) rank the top-20 world airports in number of (viable47) connections 
(as from September, 2009). Among them are six European airports: Frankfurt, Paris 
Charles de Gaulle, Munich, London Heathrow, Amsterdam Schiphol and Madrid.  
Based on the stage-length of the indirect connections, a hub may be hinterland, 
directional (hourglass) or regional.  
Hinterland hubs connect long-haul and international routes to short-haul routes. 
Hinterland hubs may be further classified based on the geographical specialization. In 
Europe some hubs are specialized for intra-European connections, so called Eurohubs, 
                                                 
45 Some measures of indirect connectivity are addressed in Chapter 3. 
46 Addressed in Section 3.3.3.  
47 It is between flights of the same airline/alliance; time between incoming and outgoing flight is between 
1h and 3h, and the detour necessary to complete the trip is not more than 20% longer than the direct 
distance between origin and destination. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
145 
 
while some other are specialized for one subgroup of routes mainly driven by social, 
economic and historical relations with the area considered, e.g. Madrid to Latin 
America. Hourglass hubs connect long-haul routes located in opposite directions. They 
are mainly unidirectional (east-west, or north-south), but with excellent geographical 
position enabling them even to be multi-directional hub. Regional hubs connect short-
haul routes.  
As a result from traffic analysis for year 2003, eight principal geographical sub-markets 
are identified at the main European hub airports (Burghouwt, 2007):  
1. Intra-European Union (EU) connections 
2. Connections between EU and other European destinations 
3. EU-intercontinental connections (five different regions are recognized48) 
4. Connections between non-European sub-markets 
In addition four categories among European hubs are recognized, (Burghouwt, 2007)49: 
1. Global all-around hubs serving long-haul to long-haul hub traffic (directional 
hubbing), short-haul to long-haul (hinterland hubbing) and short-haul to short-
haul (regional hubbing), e.g. Frankfurt, London Heathrow, Amsterdam, Paris 
Charles de Gaulle; 
2. Specialized hinterland hubs serving a specific segments of the hinterland 
hubmarket between the EU and extra-EU destinations, e.g. Munich, Lisbon, 
Paris Orly, Madrid; 
3. Regional hubs (Eurohubs and mini-hubs) serving short-haul to short-haul 
connections between intra-EU destinations (more than 65% intra-EU). From a 
geographical perspective, these connections can be EU-wide (Euro-hub) or truly 
regional (mini-hubs), e.g. Oslo, Rome Fiumicino, Helsinki, Manchester; Lyon; 
4. Directional or hourglass hubs serving the long-haul to long-haul market (extra-
EU to extra-EU). 
                                                 
48 North America, Latin America, Asia and the Pacific, Africa and Middle East 
49 Examples are updated. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
146 
 
European low-cost carriers’ networks 
LCCs have established their point-to-point networks mainly between secondary and 
regional airports that were previously underutilized or even abandoned. In addition, they 
included traditional airport where adequate potential is foreseen. Constant growth of the 
Low-cost market is achieved by: 
- Attracting new air travels, by taking over passengers from other non-competitive 
modes of transport, mainly on domestic routes,  
- Inducing higher frequency of flying of current air passengers, but also to some 
extent taking over shares of traditional airlines passengers – by operating from 
nearby secondary airports with lower fares, or from main airports with weak 
home base carriers, and  
- Taking over to large extent passengers from charter airlines – by operating from 
seasonal airports. 
Some main common characteristics of low-cost networks are summarized by 
(Dobruszkes, 2006). For example on European market LCCs operate on short and 
middle-haul routes, spreading mainly over the Western Europe. Over the past years, 
they are also strengthening west-east connections (Dobruszkes, 2009). Looking for 
niches, LCCs established their services primarily at airports not operated by national 
carriers, being at the same time poorly connected to the rest of the country or region by 
other modes of transport (primarily rail services).  
LCCs do not necessarily operate out of capital cities. In large countries there is a need 
for connecting the province with the capital city.  On the other hand, some capital cities 
(e.g. in Central and Eastern Europe) may be poorly connected to Western Europe, due 
to being served by a weak national airline. In such case LCCs can complement the 
routes.  
Relatively large low-cost networks have been developed from/to countries where 
charter flying was already successful (mainly north-south connections). LCCs do not 
systematically concentrate their demand, having the airports be regular entrance and exit 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
147 
 
points to/from the network, as opposed to being predominantly transfer/connecting 
points. 
Considering the location of low cost airport, as well as volume and share, there are five 
types of airports serving the LCCs (Dobruszkes, 2006): 
1. Medium or large traditional airports, e.g. Athens, Dublin, London Gatwick, Nice 
and Budapest; 
2. Secondary urban airports of large cities, e.g. Rome Ciampino, London Stansted, 
London Luton and Milan Bergamo; 
3. Regional airports in the wider vicinity of large cities, e.g. Hahn, Gerona and 
Liverpool; 
4. Remotely located regional airports (used by airlines to access tourist areas), 
5. Traditional airports of tourist zones, e.g. Malaga, Alicante, Faro, Palma de 
Mallorca and Porto. 
Changes/evolution of the European network 
The constantly growing LCCs market is based on successful competiveness with other 
markets and continuous opening of new opportunities. LCCs had significant influence 
on air transport networks evolution in the past, and they will most likely continue to 
have similar influence it in the future. It became evident that they did not intend to stick 
“only” to point-to-point networks they set up and strengthened over the time, but that 
they are also prepared to change their initial concept (at least partially) by: entering 
primary airports, facilitating transfers, engaging in codesharing, entering alliances and 
acquiring other airlines (de Wit and Zuidberg, 2012).  
One of the examples of LCCs entrance into primary airports resulted from their 
readiness to take advantage of opportunities and take over (fully or to certain share) 
former hub airports after they suffered “de-hubbing”. After an airport is abandoned as 
the home base by a traditional airline (due to bankruptcy, or decision to withdraw due to 
network restructuring/downsizing), some former hubs were recovered by LCCs that 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
148 
 
established their services at these airports. Redondi, et al. (2010; 2012) analyzed 
different examples of de-hubbing of the world airports (37 examples50) and their 
recovery scenarios (re-hubbed; recovered as: alliance-dominated, LCC dominated, 
unallied-dominated, or combination of these three, the so-called Battleground). The 
authors did not find any case of airport being re-hubbed by another carrier. Among 
other investigated scenarios for recovery of the former hubs, on average they did not 
show any trend towards recovery if they did not consider significant LCC entrance 
(LCC dominated or battleground lead by LCC)51. The only (European) exception is 
London Gatwick, which recovered without dominant presence of LCC. It may seem, 
however, that authors favor the impact of LCC in this paper. They did not identify any 
re-hubbed airports among European de-hubbed cases, while at the same time, in some 
other recent (even earlier date) analysis of European hubs (e.g. Deutche Bank Research, 
2006; Malighetti et al., 2008; Malighetti, et al., 2009) Zurich and, in some cases, 
Brussels are categorized as hub airports. 
Another example of LCC impact on network evolution relies on their readiness to widen 
their market in the direction of connecting traffic. LCCs try to enter primary airports to 
a greater extent (both hubs and non-hub airports of large cities), and they start to use the 
naturally created potential for developing indirect connections at their own strong basis, 
offering wider coverage to their passengers. For example (Malighetti et al., 2008) 
Skyeurope52 introduced a link between Kosice and Bratislava aiming to offer additional 
connections. It does not coordinate its flights, but it gives certain “assurance” for 
interconnection. It reimburses the ticket or provides a seat on the next flight. The largest 
European low-cost airport, Stansted, offers so called “create your own connection” 
service to take advantage of potential for (not coordinated) indirect connections 
(Malighetti et al., 2008). LCC’s “self-help” hubbing becomes possible at almost no 
extra cost for the airline. Wide network coverage, high frequencies and low fares, are 
motivating enough for customers to take advantage of random connections, even if 
LCCs do not foster this behavior (Franke 2004). Ryanair has announced plans to shift 
                                                 
50 Major de-hubbing cases in Europe: Gatwick, Brussels, Zurich, Basel, Clermont-Ferrand, Barcelona and 
Milan Malpensa (source: Redondi et al., 2010). 
51 European examples are: Birmingham (FlyBe), Basel-Mulhouse (Easyjet Switzerland), East Midlands 
Nottingham (bmibaby), Glasgow (easyJet) and Milan Malpensa (easyJet). 
52 At that time it was leading LCC in the Eastern Europe. It suffered bankruptcy in 2009. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
149 
 
their activities to major airports (excluding only three busiest, Frankfurt, London 
Heathrow and Paris Charles de Gaulle, from its plans), (De Wit and Zuidberg, 2012). 
Castillo-Manzano et al. (2012) analyzed and discussed network changes in Spain’s 
market with large presence of LCCs at hubs. At first they confirm the “common belief” 
that LCCs do not have any negative effect on traditional airlines in the domestic market, 
because they attract new air passengers who might have changed from other modes of 
transport. In the international market, they recognized three different effects – two 
supplementary (external and internal) and one complementary. On intra-European 
routes it was identified that with the increase in low-cost flights from regional and 
secondary airports, the number of flights operated by traditional airlines from main hub 
airports to the same destination had fallen to a similar extent (external supplementary 
effect). When operating the same routes (mainly intra-European, but also small number 
of non-European routes) from both secondary airports and hubs, it is evident that LCCs 
provide tough competition to traditional airlines (internal supplementary effect). On the 
other hand, the often-neglected complementary effect is identified at hub airports for 
long-haul routes. Low-cost traffic between the hubs and other national airports boosts 
international traffic with Europe (non-European destinations), and to a lesser extent with 
the rest of the world. Therefore, low-cost national flights act as feeders to traditional 
airlines for their long routes, reinforcing the role of the hub in the airport system to an 
even greater extent. It is not a usual practice but there are some examples of formalized 
agreements on collaboration, e.g. between Vueling and Iberia for feeding long haul 
flights in Madrid. The presence of LCCs at hub airports have an additional advantage 
for low-cost passengers, offering them a wider variety of connections.  
De Wit and Zuidberg (2012) list other codeshare agreements between LCCs and home 
based carriers at major hubs: WestJet and Air France-KLM; JetBlue and Lufthansa and 
Jetstar Airways and American Airlines. They even see further development towards 
LCCs starting their own hub operations, as Vueling recently did at Barcelona. 
The intensive and constant breakthroughs of LCCs into the market may be construed as 
something that will be a serious threat to the hub-and-spoke concept of full-service 
carriers, if they do not respond adequately to the changing environment (thoroughly 
discussed by Franke 2004).  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
150 
 
The present situation shows many positive effects of LCCs, while some negative ones 
may also be foreseen. Recovery of hubs after being abandoned by their home-base 
airline is certainly a positive outcome. Increased presence of LCCs at hub airports is 
also primarily positive. On one hand, LCCs provide additional passengers for traditional 
airlines on long-haul routes. On the other hand, they may offer a wider range of 
connections to their own passengers (who are willing to wait longer). From the airport’s 
perspective LCCs increase capacity utilization by operating primarily in off-peak 
periods.  
Some (at the moment isolated) examples show that LCCs are attempting to set up their 
own hub operation e.g. Vueling, as mentioned above. The other is Ryanair’s attempted 
to take over Air Lingus in order to set up connections between European LCC routes 
and long-haul to North America, through its Dublin hub (de Wit and Zuidberg, 2012). 
Another example (by same authors) to prove increased pressure by LCCs on the long-
haul market is Japanese Skymark Airlines as the first airline to order the A380.  
On the other hand, full-service carriers are protecting their long-haul markets by 
merging and operating multi-hub systems. The current state of the European network 
for long-haul routes: Air France-KLM operate from Paris CDG, Amsterdam and to a 
very small extent from Lisbon; Iberia-British Airways from London Heathrow, Madrid, 
Barcelona and to a small extent from London Gatwick; and Lufthansa Group (as the 
largest) from various airports: Frankfurt, Munich, Vienna, Zurich, Brussels, and to very 
small extent from Dusseldorf, Rome Fiumicino and Milan Malpensa (Burghouwt, 
2013). 
The hubbing itself does not seem to be endangered, as it is indispensable (at least) for 
long-haul routes. Kraus and Koch (2006) argue that “the concept of consolidating traffic 
at major airports will remain the dominant approach in the foreseeable future”. Those 
who operate might change. To survive in the competitive environment, both LCCs and 
full-service carriers will most probably need to introduce certain changes to their 
original business models (Franke, 2004). 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
151 
 
Appendix 2 – Airport categorizations accounting for hubbing 
Deutsche Bank Research (2006):  
1. Mega-hubs, through which the big, financially strong scheduled carriers and the 
strategic alliances organize their business. Given their large catchment area they 
are also attractive for point-to-point traffic (especially London and Paris), but 
are too expensive for many LCCs. 
European examples: London-Heathrow, Paris Charles de Gaulle, Frankfurt am 
Main and, to a lesser extent, Amsterdam and Madrid.  
2. Secondary hubs, which have an attractive catchment are function both as feeder 
airports for mega-hubs as well as for small partners of a strategic alliance or 
else have a hub function for certain regions. They are not overly large size in 
comparison with mega-hubs. These airports can be affected by the 
consolidation in air transport if smaller airlines using the airport as a hub are 
taken over by larger ones. At the same time, though, the free capacities are 
linked with opportunities to win new customers (scheduled, low-cost and 
charter carriers).  
European examples: Barcelona, Copenhagen, Lisbon, Manchester, Milan, 
Munich, Oslo, Rome, Stockholm, Vienna and Zurich. 
3. Secondary airports, which basically have an attractive catchment area, are 
important for feeding traffic into the big hubs and offer a certain number of 
direct scheduled connections, although intercontinental flights are the exception 
here. In principle, they are attractive also for the LCCs. The airports have no (or 
merely a rudimentary) hub function.  
European examples: Berlin-Tegel, Bilbao, Birmingham, Cologne, Dusseldorf, 
Geneva, Hamburg, Lyon, Nice, Turin and Valencia. 
Buyck, C. (2008) (cited in: EC, 2008):  
1. Intercontinental or Primary Hub. Airport with at least one based network 
carrier offering connecting opportunities. The airport offers numerous long-haul 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
152 
 
destinations (at least 20), which are not necessary all operated by the home 
carrier.  
Examples include London Heathrow, Beijing Capital, Chicago O’Hare, Dubai 
International and Mexico City. 
2. Secondary Hub. Airport with at least one based network carrier offering 
connecting opportunities. The airport offers several intercontinental routes 
and/or numerous medium-haul routes.  
Examples include Denver, Istanbul, Brussels, Dublin and Amman. 
3. Regional platform. Airport that is not a hub and thus traffic is mainly point-to-
point. The airport traffic is focused on short/medium-haul routes.  
Examples include Barcelona, Cancun and Fukuoka. 
4. LCC base. Airport with at least one based LCC. International hubs and regional 
platforms all can be an LCC base as well. 
Burghouwt, G. (2007): 
1. 1st tier airports - are the primary home bases of carriers that operate on a global 
or continental scale. The home carrier may or may not (due to saturation) 
operate a wave-system structure at such an airport. 
2. 2nd tier airports - are somewhat smaller airports in terms of the indirect 
connections generated compared with the 1st tier airports. These airports are the 
home bases of at least one home carrier that operates a specialized hinterland, a 
European, or directional hub/traffic node. A wave-system structure may or may 
not be present here. 
3. 3rd tier airports - Most of them are the home base of at least one airline 
operating a European traffic node or hub. Apart from a few exceptions, no wave-
system structure can be found at these airports. They are used as secondary or 
tertiary nodes in networks of large airlines and as primary nodes in the networks 
of smaller airlines.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
153 
 
4. 4th tier airports - home based airlines offer only very few indirect connections. 
Most of them are used as secondary or tertiary nodes in airline networks. No 
wave-system structure is present. 
5. 5th tier airports - are the smallest airports in the airport hierarchy and do not 
offer any significant amount of indirect connections. They do not have an 
intermediate function in the aviation network, but are the terminal destinations 
for virtually every air passenger. For a hub-and-spoke airline, such airports may 
function as spokes in the route network. 5th tier airports cover a wide range of 
airports. 
However, this categorization is based on weighted indirect connection number (WNX) 
derived based on traffic statistics from the year 1999. It does not reflect current state 
which is confirmation of constantly evolving air-transport networks. 
Malighetti et al. (2009) clustered European airports in eight clusters, as follows: 
1. Worldwide hubs (8 airports) - Airports served by worldwide alliances. They 
have similar dimensional characteristics, especially in terms of density and 
intensity. They have a high percentage of overseas destinations and can offer a 
wide range of opportunities for interconnectivity.  
Examples: London Heathrow, Paris Charles de Gaulle, Amsterdam, Frankfurt, 
Rome Fiumicino. 
2. Hubs (16 airports) - Airports are mainly former flag carrier hubs and secondary 
medium or large-size hub airports. These airports tend to favor European routes 
(on average 75% of the routes are within Europe) and are served by a variety of 
LCCs.  
Examples: Athens, Vienna, Zurich, Brussels, Stockholm Gatwick, Orly, Milan 
Malpensa, London Stansted. 
3. Secondary gates (11 airports) – Airports of medium-size dimensions and offer a 
limited number of overseas destinations (less than 30% of the routes offered), 
with a visible concentration of traffic distribution over a limited number of 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
154 
 
destinations. Their role as intermediate connections further confirms their main 
function as gates for local areas.  
Examples: Lisbon, Glasgow, Venice, Warsaw and Marseille. 
4. Airports with 3-5 million passengers per year (33 airports). They have a high 
concentration of LCCs (covering 75% of the seats offered, on average). The 
main destinations are generally non-domestic and European (only 15% of the 
destinations are domestic, against an average of European destinations higher 
than 90%).  
Examples: Bergamo Orio al Serio, Ciampino and Pisa. 
5. No low-cost gates (46 airports). The term “gate” here refers to the European 
rather than overseas market. This cluster seems to be less homogeneous than all 
the previous clusters. Destinations here are almost always European. LCCs play 
a minor role and routes are mainly domestic and intended to connect secondary 
airports within the country. Some airports may be connecting points between 
hub and secondary airports, as can be seen from the average of the limited % 
index (up to 63%) and from the existence of a number of heavily flown routes 
along with less demanded flights (the index of distribution of the routes is higher 
here than in any other cluster).  
Examples: Milano Linate, Palermo and Valencia. 
6. Regional airports (44 airports). This cluster is made of smaller, mainly regional 
airports. As is the case with the previous cluster, it is not very compact. 
7. Airports for LCCs mainly (71 airport). This cluster consists of airports 
dimensionally similar to cluster 6, but mainly featuring a large presence of 
LCCs, determining a wide offer of European rather than domestic destinations. 
8. Local airports (238 airports). This is the largest cluster and is made up of local 
airports. Their offer is often limited to a restricted number of routes, only rarely 
more than 4 or 5, generally touching domestic destinations; LCCs are seldom 
found to operate here and are absent in more than 75% of the airports listed in 
this category.  
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
155 
 
Airport categorization by Munich Airport (cited in: Oettl and Boeck, 2012): 
Airport type Examples Characteristics 
International Hubs FRA, LHR Transfer hub for intercontinental traffic; 
Global service area; 
Major node for alliances; 
More than 40 million pax per year (2008). 
Secondary Hubs MUC, VIE Hub with mainly traffic within Europe; 
Service area mainly concentrated to Europe, partially global 
possible; 
Secondary node for alliances; 
more than 15 million pax per year. 
International O/D TXL, GVA Traffic mainly within Europe; 
Regional service area; 
Few long-range destinations; 
More than 5 million pax per year. 
Regional O/D SZG, TRD Feeding traffic to hubs, several direct destinations and/or low-cost 
destinations; 
Local service area; 
Less than 5 million pax per year. 
Secondary O/D HHN, OLB Signifficant low-cost traffic; 
Produst differentiation against other airports; 
Local to regional service area. 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
156 
 
Appendix 3 – European seasonal airports  
Typical European seasonal airports representatives (source: Eurocontrol, 2007) are 
listed in Table A3-1. 
Table A3-1. European seasonal airports in 2006 
 
 
ICAO 
code Country 
Median  
Depart. 
Airport 
size  
(Thousand 
Annual 
DEPs) 
Peak 
date 
(2006) 
Week 
day 
Max 
Dep./ 
Med 
Dep. 
Peak 
Description 
Ioannis/ 
Kapodistrais LGKR Greece 11 5 – 10 Aug.26 Sat 8.4 W&S* 
Burgas LBBG Bulgaria 9.5 5 – 10 
Jul.25 & 
Aug.8 Tue 7.9 W&S  
Split LDSP Croatia 15 5 – 10 
Jul.22 & 
Aug.12 Sat 6.3 W&S 
Kos LGKO Greece 11 5 – 10 Aug.2 Wed 6.2 W&S 
Mugla-
Dalaman LTBS Turkey 17 5 – 10 Aug.28 Mon 6.2 W&S 
Milas/Bodrum LTFE Turkey 16 5 – 10 Aug.14 Mon 6.0 W&S 
Varna LBWN Bulgaria 11 5 – 10 
Jul.21 & 
Aug.11 Fri 5.8 W&S 
Cannes 
Mandelieu LFMD France 17 5 – 10 May.29 Mon 4.5 Spec. event 
Dubrovnik LDDU Croatia 16 5 – 10 Jun.25 Sun 4.2 W&S 
Tempelhof-
Berlin EDDI Germany 43 10 – 20 Jul.10 Mon 6.8 Spec. event 
Olbia Vosta 
Smeralda LIEO Italy 26 10 – 20 Jul.23 Sun 5.2 W&S 
Diagoras LGRP Greece 30 10 – 20 Jul.16 Sun 4.8 W&S 
Mahon/ 
Menorca LEMH Spain 31 10 – 20 Aug.4 Fri 3.5 W&S 
Innsbruck LOWI Austria 26 10 – 20 Feb.25 Sat 3.1 W&W ** 
Salzburg LOWS Austria 43 10 – 20 Feb.18 Sat 3.0 W&W 
Madeira LPMA Portugal 25 10 – 20 Jan.2 Mon 2.9 Spec. event 
Ibiza LEIB Spain 49 20 – 50 Aug.20 Sun 4.4 W&S 
Schoenefeld-
Berlin EDDB Germany 79 20 – 50 Jul.10 Mon 3.0 Spec. event 
Nikos/ 
Kazantzakis LGIR Greece 54 20 – 50 Aug.4 Fri 2.8 W&S 
Paris le 
Bourget LFPB France 85 20 – 50 May.18 Thu 2.7 Spec. event 
Antalya LTAI Turkey 119 20 – 50 Aug.5 Sat 2.5 W&S 
Arrecife 
Lanzarote GCRR Spain 59 20 – 50 Oct.26 Thu 2.2 
Weekly 
seasonality 
Palma de 
Mallorca LEPA Spain 258 50 – 100 Aug. 5 Sat 2.0 W&S 
Nice LFMN France 187 50 – 100 May. 29 Mon 1.7 Spec. event 
Manchester EGCC UK 306 100 – 200 Aug. 25 Fri 1.3 W&S 
 
* W&S – Weekly and summer seasonality 
** W&W – Weakly seasonality, stronger in winter 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
157 
 
The share of the monthly number of departures in the number of annual departures is 
given in Figure A3-1, for seven seasonal AENA airports (Mirkovic et al. 2013).  
 
Figure A3-1 - Monthly share in total number of departures in 2011 compiled from AENA Estadisticas 
(Mirkovic et al. 2013) 
These airports fall into different size categories: Palma de Mallorca and Malaga – 10 to 
25 million passengers; Alicante and Ibiza – 5 to 10 million passengers; Menorca, 
Murcia and Reus – 1 to 5 million passengers. There is a clear difference between high 
season (usually from the 4th/5th to the 9th/10th month of the year) and off-season traffic. 
Depending on the airport, the difference is smaller (Alicante and Malaga) or larger 
(Palma de Mallorca, Murcia and Reus), or even extremely pronounced in some cases 
(Ibiza and Menorca). 
  
0
0,02
0,04
0,06
0,08
0,1
0,12
0,14
0,16
0,18
0,2
1 2 3 4 5 6 7 8 9 10 11 12
sh
ar
e 
of
 m
on
th
ly
 D
EP
s 
in
 a
nn
ua
l D
EP
s
month
Alicante
Ibiza
Malaga
Menorca
Murcia
Palma de Mallorca
Reus
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
158 
 
Appendix 4 – Minimum connecting times – examples at the world’s airports 
AMS - Amsterdam Schiphol ATL - Hartsfield-Jackson  
 
Domestic to Domestic - 25 minutes  
Domestic to International/Europe - 50 minutes  
Domestic to Europe - 40 minutes  
International to Domestic - 50 minutes 
Europe to Domestic - 40 minutes  
International to International - 50 minutes  
Within Europe - 40 minutes  
 
 
Domestic to Domestic - 55 minutes  
Domestic to International - 1 hour  
International to Domestic - 1 hour 30 minutes  
International to International - 1 hour 30 
minutes  
 
BKK - Bangkok International  BCN - El Prat  
 
Domestic to Domestic - 30 minutes  
Domestic to International - 2 hours  
International to Domestic - 2 hours  
International to International - 1 hour 15 
minutes  
 
 
Domestic to Domestic - 30 minutes  
Domestic to International - 45 minutes  
International to Domestic - 45 minutes  
International to International - 45 minutes  
 
BOS - Logan International  ORD – Chicago O'Hare International 
 
Domestic to Domestic - 40 minutes  
Domestic to International - 1 hour  
International to Domestic - 1 hour 30 minutes  
International to International - 1 hour 15 
minutes  
 
 
Domestic to Domestic - 50 minutes  
Domestic to International - 1 hour 15 minutes  
International to Domestic - 1 hour 30 minutes  
International to International - 1 hour 30 
minutes  
CPH - Copenhagen  DFW - Dallas/Ft Worth International  
 
Domestic to Domestic - 30 minutes  
Domestic to International - 45 minutes  
International to Domestic - 45 minutes  
International to International - 45 minutes  
 
 
Domestic to Domestic - 50 minutes  
Domestic to International - 50 minutes  
Domestic to Canada - 50 minutes  
International to Domestic - 1 hour 10 minutes  
Canada to Domestic - 50 minutes  
International to International - 1 hour 10 
minutes  
 
Source: Airtravel about (2013)  
Munich Airport (2013):  within T2 - 30 minutes  
Vienna Airport (2013):  for Star Alliance flights - 25 minutes  
Timisoara Airport (Carpatair, 2013): Carpatair flights - 20 minutes  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
159 
 
Appendix 5 – Indirect connectivity indices 
Dennis (1994) calculated connectivity ratio, as the ratio between the total number of 
viable connections offered at the airline hub during a typical airline operational day (15 
hours) and approximate number of viable connections that would be expected in the 
case of a random timetable. A viable connection is considered the one that fulfills 
criteria of MCT of 45min and MaxCT of 90min. If scheduling is not better than random 
overall, the ratio is 1.  
Danesi (2006) proposed a weighted connectivity ratio, by combining Dennis’ (1994) 
and Burghouwt and de Wit’s (2005) approaches. He introduced intermediate connecting 
time (ICT), which is between MCT and MaxCT. For all possible arrival-departure 
connections at an airline hub during the day, temporal and spatial weights are assigned. 
In temporal connectivity matrix weight 1 is assigned to all connections with transfer 
time between MCT and ICT, weight 0.5 if transfer time is between ICT and MaxCT, 
and weight 0 if transfer time is longer than MaxCT. Similarly, in space connectivity 
matrix weights (0, 0.5 and 1) are assigned to all connections based on the de-routing 
index (ration between great circle distance of the direct route and great circle distance of 
the indirect route). Multiplying temporal and spatial connectivity matrices gives a 
weighted connectivity matrix. Summarizing all values from the weighted connectivity 
matrix gives a weighted number of viable connections offered by the airline hub during 
a typical airline operational day. The approximate number of viable connections in the 
case of purely random timetable is calculated somewhat differently, as they include ICT 
in addition to MCT and MaxCT. Their ratio finally yields the weighted connectivity 
ratio. 
Malighetti et al. (2008) suggested an approach to estimate the potential for indirect 
connectivity and analyzed how much of it was already used by the main global 
alliances: One World, SkyTeam and Star. An approach is based on the connectivity 
index (calculated as the average of the minimum paths between a certain airport and all 
other airports in the network). In order to take in account centrality of the airport, 
“betweenness” (the number of minimal paths within the entire network to pass through 
a certain airport) and “essential betweenness” (the number of unavoidable minimal 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
160 
 
paths passing through a certain airport) are determined. In addition, the average 
minimum travel times, based on flight time and waiting time (fewest possible 
connections), is determined, assuming a minimum connecting time of 60min and no 
constraints on maximum connecting time (it is reasonable considering they investigate 
potential).  
The extent potential indirect connectivity is used by the alliances, based on the average 
number of fastest paths passing through each of 20 best connected airports (all 
connections, two-step connections, as well as connections fulfilling the condition of 
connecting time between 1h and 3h are considered) was analyzed. Authors stressed that 
possible indirect connectivity is only partially exploited by alliances (2/3 remain 
unexploited). 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
161 
 
Appendix 6 – Evolution of wave-system in 1990s: Example of Munich Airport 
Burghouwt (2007) applied a theoretical approach to identify presence of wave-system 
structures at European hub airports. Hypothetical ideal wave is identified based on 
minimum connecting time of 40min and maximum acceptable connecting time of 
90min. The wave center (C) moves forward in steps of six minutes and the number of 
flights within arrival (C-45, C-20) and departure (C+20, C+45) windows for the specific 
wave center are counted respectively. An actual wave centre (C) is identified when the 
maxima of the waves for arriving and departing flights (almost completely) coincide.  
An example illustrating wave-structure evolution during the 1990s is shown in Figure 
A6-1. The number of waves the airline operates and the number of aircraft per wave are 
clearly visible.  
 
Figure A6-1. Wave-structure evolution at Munich Airport, Germany (Burghouwt , 2007) 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
162 
 
However, with this approach, information about the exchange between the arrival time 
window and the departure time window is somehow lost. They appear as overlapped, 
which is not the case. Traffic patterns at (Lufthansa’s) Terminal 2 at Munich Airport, 
for summer 2000 and 2005 are given in Figures A6-2 and A6-3, respectively (source: 
Munich Airport53). 
 
Figure A6-2. Lufthansa’s wave-system structure at Munich Airport Terminal 2, summer 2000 
 (source: Munich Airport) 
 
Figure A6-3. Lufthansa’s wave-system structure at Munich Airport Terminal 2, summer 2005  
(source: Munich Airport) 
                                                 
53 Provided through the courtesy of Michael Hoehenberger, Head of Operational Planning at MUC 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
163 
 
Appendix 7 – Selected definitions of peak hour and design day 
Selected definitions of Peak Hour (de Neufville and Odoni, 2002; Ashford et al, 2013): 
- Standard busy rate - the 30th busiest hour of the year; also 20th or 40th busiest 
hour of the year, 
- Typical peak hour (passengers) - the peak hour of the average day of the peak 
month of the year; also the peak hour of the average day of the two peak months 
of the year, 
- The peak hour of the 95th percentile busy day of the year, i.e. the peak hour of, 
roughly, the 18th busiest day of the year; also the peak hour of the 7th or 15th 
busiest day of the year, 
- Busy-hour rate - the “5 percent busy hour,” i.e., an hour selected so that all the 
hours of the year that are busier handle a cumulative total of 5 percent of annual 
traffic, 
- The peak hour of the 2nd busiest day during the average week in a peak month, 
- Peak profile hour or average daily peak – largest hourly volume in the average 
peak day of the peak month,  
- Busiest timetable hour – calculated from average load factor and 
existing/projected timetable (for small airports with limited data bases). 
The U.S. FAA uses peak hour of the average day of the peak month of the year. The 
ICAO recommends the peak hour of the average day of the two peak months of the year 
(de Neufville and Odoni, 2002). In the United Kingdom (U.K. BAA – British Airports 
Authority) before busy-hour rate was introduced, standard busy rate (mainly) and peak 
profile hour (in some cases) were used. Aeroports de Paris use 3 percent busy hour. 
Dutch airports use the sixth busiest hour, which corresponds to average of the 20 
highest hours (Ashford et al, 2013). 
ACRP report (TRB, 2013) stated that the most common current practice in the U.S. is to 
define the design day as peak month average day. It is calculated by identifying the 
month with the highest number of operations and/or passengers, and then dividing the 
operations or passengers in that month by the number of days in the month. There are 
several other design day definitions in use: 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
164 
 
- The average week day in the peak month, 
- The 15th busiest day of the year, 
- The 30th busiest day of the year, 
- The 90th percentile—corresponds to the 36th busiest day of the year. 
Peak month average day is easy to calculate but it can generate very different design 
day thresholds from airport to airport, ranging from 20th or 15th busiest day of the year, 
at an airport with high seasonality to even 100th or 150th busiest day of the year, at an 
airport with low seasonality, especially one with some day of the week variation in 
activity.  
ACRP report discusses several ways of estimating future design day profiles by:  
1. Assuming the base year distribution of daily activity will carry forward 
unchanged into the future.  
2. Assuming the peak spreading component based on relationships between airport 
size and peak period percentage. This dampens the peaks and fills in the gaps in 
the daily schedule.  
3. Generating daily profiles by category of activity (i.e., domestic and international 
passengers), project each profile to grow at the annual rate of the corresponding 
activity category, and then aggregate the results to generate an estimated future 
daily profile.  
4. Aggregating a daily profile from a design day schedule. 
On the other hand, design day schedule is prepared by modifying an existing schedule 
to include assumptions on new markets, additional frequencies, and fleet mix changes. 
In some instances daily profiles are derived from design day schedules. In other 
instances, previously derived daily profiles are used to guide the addition of flights for 
future design day schedules. The design day forecast provides control totals for 
passengers and aircraft operations. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
165 
 
Appendix 8 – Custer analysis of the world’s airports with respect to traffic 
parameters  
Oettl and Reeb (2012) analyzed 203 airports around the world with more than 18 
operations per hour. Seven consecutive days of scheduled operations data (obtained 
from: OAG, 2008) were taken into consideration. In the clustering process each day is 
treated as a separate airport, since hourly traffic distribution may significantly differ 
between the days of the week. Representative peak periods and type of airports with 
respect to traffic parameters are shown in Figure A8-1 and Figure A8-2, respectively. 
 
Figure A8-1. Representative peak periods (Oettl and Reeb, 2012) 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
166 
 
 
Figure A8-2. Airport categories with respect to traffic parameters (Oettl and Reeb, 2012) 
In addition Figure A8-3 gives the results from a trial to categorize airports with respect 
to additional traffic parameters related to users and the type of traffic (Oettl and Boeck, 
2012).  
The parameters selected to describe category of European hubs are:  
− available seats to destinations within Europe (ASE);  
− share of hub carrier alliance(s) (HAL);  
− aircraft size described through share of heavy aircraft (HVY);  
− flight distances: Short Range (SR≤1500 km), Medium Range (1500 km <MR< 
3000 km), Long Range (LR> 3000 km);   
− share of airline types: Home Carrier (HC), Flag Carrier (FC), Intercontinental 
Carrier (IC), Low Cost and Charter Carrier (LC). 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
167 
 
 
Figure A8-3. Traffic parameters for selected group of European hubs (Oettl and Boeck, 2012) 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
168 
 
Appendix 9 – Models for optimizing available resources utilization 
Before making huge infrastructural investments to expand their existing capacity, 
airports can do something to increase efficiency of capacity utilization.  
In the past 20 years, when it was obvious that airports were becoming the most critical 
area in the air transport network in terms of capacity-demand imbalance, many decision 
support tools for available (limited) airport resources management have been developed, 
tested and introduced in practice. As expected, the first tools were developed for runway 
system capacity management, since the major bottlenecks on the airside come from the 
runways.  
It began with AMANs (Arrival Manager-s), which were developed to support arrival 
sequencing in order to increase runway throughput and avoid generation of delays in 
peak periods (e.g. Beasley et al. (2000); Venkatakrishnan et al. (1993); one of the 
commercial solutions: CALM - Computer-assisted Approach and Landing 
Management, developed by Barco, for Zurich Airport; Barco (2004)). Then, a series of 
DMANs (Departure Manager-s) were developed, for sequencing departing aircraft in 
order to minimize waiting time and queue lengths (e.g. Anagnostakis et al. (2000); 
Boehme (2005); Feron et al. (1997); Hasselink and Basjes (1998); Jonge et al. (2005)). 
Coordination between AMANs and DMANs followed (e.g. Boehme et al. (2007)), as 
well as the development of SMANs (Surface Managers) (e.g. Atkins and Brinton 
(2002); Lowson (1997)), for surface movement optimization and TMANs (Turnaround 
Managers) for turnaround process optimization (one of the commercial solutions: GS 
Hub Control, Inform (2013)). 
The general tendency is to integrate all those managers into a single overall manager 
aimed at optimization of available resources utilization or exploitation from the 
viewpoint of all stakeholders throughout (Airport) Collaborative Decision Making (e.g. 
Guenther et al. (2006) describing the operational concept and logical architecture of 
TAM - Total Airport Management; Pick (2007), describing test results of CLOU - Co-
operative Local resOUrce planner, at Frankfurt Airport; Pina et al. (2005), presenting 
research results on LEONARDO - Linking Existing on Ground, Arrival and Departure 
Operations, tested at Madrid Barajas Airport). 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
169 
 
Based on the existing tools, it is clear that the optimization of airport airside utilization 
is mainly focused on the runway system. Not only on arrivals and departures managers, 
but other optimization tools (optimization of surface movements and the turnaround 
process) also aimed at achieving better runway system utilization. They are locally 
focused on certain airside elements, but, when integrated, they have a common goal 
function, which is runway performance improvement. There is no such optimization 
tool focused on improving the efficiency of apron resources (stands and gates) 
utilization. At least, not the ones that observes aprons as part of the airside.  
The majority of apron optimization tools observe the apron (precisely contact stands i.e. 
gates) from the perspective of the terminal complex (landside). 
A very common, and widely addressed, optimization problem in the area of apron 
utilization optimization is the gate assignment problem (optimization of gate/stand 
utilization). Many papers deal with this issue (some of the earliest: Babic et al. (1984), 
Mangoubi and Mathaisel (1985), Hamzawi (1986); and one of latest: Genc et al. 
(2012)). The majority of them are based on the minimization of passenger walking 
distances from check-in to gate and from gate to baggage claim area, as well as from 
gate to gate for transfer passengers. Cheng (1997); Haghani and Chen (1998); Dorndorf 
et al. (2007) give a thorough literature overview on research results in this area, 
classified by methods, goal functions, etc. These models result in aircraft stand 
assignment schemes considering the contact stands (gates) from the perspective of the 
landside. The objective is to improve and maintain a certain level of service in the 
terminal building. 
Ding et al. (2005) expand the gate assignment problem by considering over-constrained 
cases, where the number of aircraft exceeds the number of available gates. They address 
both the objectives of minimizing the number of ungated aircraft and minimizing total 
walking distances.  
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
170 
 
Appendix 10 - Comparison of the analytical model and FAA graphical method 
For quick estimation of apron-gate capacity U.S. FAA has proposed graphical method 
(FAA, 1983). An example from AC 150/5060 is used and the results from graphical 
approach are compared to results from the extended analytical model. 
Example 6, from AC 150/5060: An apron has 10 stands allocated to three airlines, X, Y 
and Z. Apron X has 4 stands for small (narrow-body) aircraft and 1 stand for large 
(wide-body) aircraft, apron Y has 2 stands for small and 1 for large aircraft and apron Z 
- 2 stands for small aircraft. During an hour, airline X schedules 13 small aircraft with 
an average stand occupancy time (SOT) of 45min, and 2 large aircraft with an average 
stand occupancy time of 55min. Airline Y schedules 8 small (SOT=40min) and airline 
Z 4 small aircraft (SOT=35min).  
When the numbers of flights are expressed by the shares in total demand, as given in 
Table A10-1 (Column 4), extended apron capacity model provides the following results:  
Table A10-1. Demand structure and average stand occupancy times in Example 6 
 
( )31211211 ,,,min CCCCC =  
⎟⎠
⎞⎜⎝
⎛ ⋅⋅⋅⋅
+⋅⋅⋅⋅+⋅
+= 60
3515,0
2,60
403,0
12,60
5507,0
1,60
5507,04548,0
14minC
 
( ) haircraftC /8,118,22;15;6,15;8,11min ==   
Apron capacity is limited by the first (of four) group of stands. 
The FAA’s graphical method, shown in Figure A8-1, estimates apron-gate capacity 
expressed in movements/h as: NSG ⋅⋅* , where: *G is the hourly gate capacity base, S 
is the gate size factor and N is the number of gates. 
number of 
stands user
aircraft 
class
share in 
population (%)
avg. stand 
occupancy time 
(min)
4 1 48 45
1 2 7 55
2 1 30 40
1 2 0 0
2 airline Z 1 15 35
airline X
airline Y
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
171 
 
  
Figure A10- 1.  FAA’s graphical method for calculating gate capacity (FAA, 1983) 
For the given example (Table A10-2) apron-gate hourly capacity is: 13 movements/h for 
apron X, 9 movements/h for apron Y and 7 movements/h for apron Z. Overall apron-
gate capacity is estimated (incorrectly) as a sum of these three capacities which makes 
29 movements/h.  
Table A10-2. Parameters derived for graphical approach in Example 6 
 
In order to compare the results from FAA’s graphical method and analytical model, let 
us observe aprons X, Y and Z separately. It makes the following: 
For apron X: haircraftCx /5,6605513,0
1,60
5513,04587,0
5min =⎟⎠
⎞⎜⎝
⎛ ⋅⋅⋅⋅+⋅=  
Gate 
occupa
ncy 
ratio
Hourly 
capacity 
base
Gate size No.Gates Hourly capacity
N W N W N (%) W (%) N (Tn) W (Tw) (Tw/Tn) G* S N G* S N
X 13 2 4 1 80 20 45 55 1,22 2,6 0,97 5 13
Y 8 0 2 1 67 33 40 0 1,00 3 1,00 3 9
Z 4 0 2 0 100 0 35 0 1,00 3,4 1,00 2 7
29
No. Gates Gate mix Average gate time
capacity of the terminal:
Demand
Gate 
group
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
172 
 
(The share of wide-body/non-wide-body aircraft if only flights of airline X are 
considered, is 87/13 %.)  
For apron Y: haircraftCy /5,460401
3 =⋅⋅=   
For apron Z: haircraftCz /4,360351
2 =⋅⋅=   
(Airline Y and Z operate all flights, 100%, with non-wide-body aircraft.) 
Capacities set by each apron calculated from the graphical method (expressed in 
movements/h) are double than values calculated analytically (expressed in aircraft/h). 
As a quick approximation, one can multiply the dynamic capacity of the apron by two to 
convert it to movements per hour, as the occupancy of a stand is associated with two 
movements on the runways, an arrival and a departure (De Neufville and Odoni, 2003). 
If this approach is applied, then the same result is obtained as by FAA’s graphical 
method (for each apron).  
Nevertheless, the final result from the graphical method leads to the wrong conclusion 
as it overestimates capacity of the entire apron. As discussed before, capacities of 
individual areas should not be summed. Each area restricts overall apron capacity to the 
certain level, depending on mix of users in demand. Capacity restriction set by each 
apron has to be calculated respecting the share of each airline in total demand 
(respectively 55%, 30% and 15% for airline X, Y and Z) and minimum adopted as total 
apron capacity. This implies that overall capacity of the apron is not: 
haircraftCCCC zyx /8.144.35.45.6 =++=++=  
but it should be determined as: 
( ) haircraftCCCC zyx /8,118,22;15;8,11min
15,0
,
3,0
,
55,0
min ==⎟⎟⎠
⎞
⎜⎜⎝
⎛=  
which is the same result as obtained analytically.  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
173 
 
Appendix 11 – Apron with flexible layout with respect to aircraft classes 
For estimating capacity of the apron with flexible contact stands area, in the first step 
several typical (most often) layout configurations have to be defined, e.g. I, II and III.  
Capacity is calculated for each of them individually (like they are separate aprons).  
( ) ⎟⎟⎠
⎞
⎜⎜⎝
⎛==
'
'
minmin
ji
ij
ij
I
ijI t
N
CC
ij
 
( ) ⎟⎟⎠
⎞
⎜⎜⎝
⎛==
''
''
minmin
ji
ij
ij
II
ijII t
N
CC
ij
 
( ) ⎟⎟⎠
⎞
⎜⎜⎝
⎛==
'''
'''
minmin
ji
ij
ij
III
ijIII t
N
CC
ij
 
Apron capacity is maximum capacity provided by different apron configurations:  
( )IIIIII CCCC ,,max=
 
It is the one which to a highest degree matches to demand structure with respect to 
aircraft classes. 
 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
174 
 
Appendix 12 – Ideal wave with intercontinental and continental flights 
The ideal wave accounting for both intercontinental and continental flights is given in 
Figure A12-1 (Bootsma, 1997; cited in: Burghouwt (2007)).  
 
Figure A12-1. Ideal wave involving both continental and intercontinental connections (Bootsma, 1997, 
cited in: Burghouwt, 2007) 
Where: 
C – represents the wave center,  
Tc – the maximum acceptable connecting time for continental connections,  
Ti/c – the maximum acceptable connecting time involving intercontinental 
connections,  
Mc – the minimum connecting time for continental connections, and 
Mi/c – the minimum connecting time involving intercontinental connections. 
 
The same wave is given in Figure A12-2, in the form used in the thesis.  
 
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
175 
 
 
 
Figure A12-2. Ideal wave involving both continental and intercontinental connections 
In this case we have: 
A
t
MCTMaxCTN
CC
CICC
los ⋅
−=+
2
/
 
A
t
MaxCTMaxCTN
CIC
IC
los
−=  
IC
los
ICCC
loslos NNN += + /  
}{ losca NNN ,min /=  
STTATSOT +=  
CC
tw
C MCTARRTAT +=  
ICCICC
tw
CIC MCTARRTAT += + //  
CIC
tw
IC MaxCTARRTAT +=  
1 2 3 4 6 7 8 9
IC IC/C C
C IC/C IC
MCTC
MCTIC
MaxCTC
MaxCTIC
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
176 
 
Appendix 13 – An impact of the runway system performance on apron capacity 
Example 7 assumes the minimum separation between all aircraft classes on approach is 
6Nm, and the minimum separation between all departing aircraft is 120sec. The apron is 
of the same structure as in Example 3 - 22 contact stands (of which 12 for aircraft class 
2, and 10 for aircraft class 3) and 8 remote stands (of which 5 for aircraft class 1, and 3 
for aircraft class 2). Hubbing parameters are the same as in Example 3a - WRC is 
180min; MCT is 30min and MaxCT is 150min.  
The demand structure is the same as given in Table 9 (Table A13-1). A modified O/D 
example is used to be comparable to hub with mix of coordinated and other (point-to-
point) flights. 
Table A13-1. Demand structure for Example 7 
 
Runway capacity decreases to 21 arrivals/h, 41 operations/h (alternating arrivals and 
departures) and 30 departures/h. The impact of runway performance on apron capacities 
is shown in Figure A13-1.  
The blue and red lines represent runway capacity (arrivals/h) in mix mode, for Example 
3 and Example 7 respectively. In Example 3 airside capacity is limited by the runway 
capacity for the O/D airport, and by apron capacity for hub airport (all cases). In 
Example 7 airside capacity is limited by the runway capacity for all airport types.   
In the case of an O/D airport, apron capacity remains the same regardless of runway 
capacity; the consequent mismatch between the two is higher in Example 7 (40 vs. 21 
arrivals/h) than in Example 3 (40 vs. 33 arrivals/h). 
flight type aircraft class HUB pure
HUB 
exclusive/
preferential
O/D
1 0,2 0,08 0
2 0,6 0,24 0
3 0,2 0,08 0
1 0 0,36 0,44
2 0 0,24 0,48
3 0 0 0,08
coordinated
other
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
177 
 
 
Figure A13-1. Apron capacity sensitivity to runway system performance, comparison of Example 3 and 
Example 7 
At hub airports, the influence of runway capacity on apron capacity can be seen in all 
cases (except the case of exclusive use).  
In Example 7, Nlos (20 aircraft) becomes more constraining than static apron capacity 
(Na/c=22 aircraft). Also, apron capacity at the hub airport changes.  
As it can be seen in Figure A13-1, theoretical and utilized capacity of the hub airport 
serving only coordinated flights is somewhat lower. 
The impact of the decrease in the runway performance on apron capacity can also be 
seen for hub airports serving mixed coordinated and point-to-point traffic in the 
preferential use case. There, it is constrained by the (utilized) capacity of the group of 
stand for coordinated flights. In the exclusive use case, the influence of the runway 
performance is not visible since apron capacity is limited by the capacity of the group of 
stands for other flights/aircraft class 2.  
  
0
5
10
15
20
25
30
35
40
45
O/D ‐ all Static 
capacity
LOS 
capacity
HUB ‐
theoretical
HUB ‐
utilized
MIX ‐
exclusive
MIX ‐
preferential
ap
ro
n 
ca
pa
ci
ty
 (a
ir
cr
af
t/
h)
Example 3 Examle 7
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
178 
 
Appendix 14 – Apron capacity estimation tool – basic user guide 
An apron capacity estimation tool was developed as a supplementary material to this 
thesis. It is available at: http://apron.comze.com/ (login password: bojana).    
The tool offers apron capacity estimation for origin-destination (O/D) airports and for 
hub airports. It contains the following input data:  
1. Number of airport users, 
2. Number of aircraft types, 
3. Total number of contact and remote stands, 
4. Hubbing parameters: arrival only runway capacity, departure only runway 
capacity, arrival runway capacity that allows one departure between two arrivals, 
minimum connecting time, maximum acceptable connecting time, and wave 
repeat cycle, 
5. Number of contact stands by user and aircraft class, 
6. Number of remote stands by aircraft class, 
7. Rules on contact stands use, 
8. Rules on remote stands use, 
9. Turnaround time by user and aircraft class, 
10. Separation time by user, 
11. Demand structure by user and aircraft class,  
12. Coordinated/other flights share – derived from 11, only for hub case. 
Inputs 4 and 12 apply only for hub airports. All other inputs are common for O/D and 
hub airports. 
For O/D airports the tool delivers the following outputs: 
1. Apron capacity of contact stands only, and 
2. Apron capacity of all stands (contact + remote). 
For HUB airports it delivers the following outputs: 
1. Maximum number of aircraft in a wave, and 
2. Apron capacity. 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
179 
 
 
Figure A14-1. Apron capacity estimation tool - input data 
For O/D airports output 1 and output 2 provide information about gate capacity and 
additional capacity when remote stands are engaged, respectively. Output 2 is useful for 
airports with significant variations in daily, weekly and yearly traffic, where remote 
stands may be used to deal with traffic overflow. When analyzing additional capacity 
provided by remote stands the user should carefully define the rules on stand usage in 
order to catch desired relations. For example, remote stands should be assigned as 
additional resources only to one user in the exclusive use case, while in the preferential 
or common use cases remote stands can be allowed for various users that share contact 
stands at the same time.  
 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
180 
 
 
Figure A14-2. Apron capacity estimation tool - output for O/D airports 
For hub airport cases, the assumption is that coordinated flights use contact stands (in 
order to facilitate transfers more efficiently), while other flights do not have their own 
contact stands, i.e. number of contact stands for other flights should be set as zero. 
Other flights use remote stands only, in the exclusive use case. In addition to remote 
stands, other flights also use contact stands of coordinated flights in the preferential use 
case. 
For hub airports output 1 provides information on the maximum number of coordinated 
aircraft that can be scheduled within a wave. It takes into consideration static apron 
capacity and constraints imposed by the level of service in passengers’ transfer and 
runway system performance (given in hubbing parameters input).  
Output 2 delivers apron capacity for pure hub (all flights are coordinated) and for hub 
with mixed coordinated and other (non-coordinated) flights in demand. The difference 
between exclusive use and preferential use cases is controlled by the rules on contact 
stands use. For the preferential use case, there is also the possibility of analyzing apron 
capacity when coordinated flights use additional remote stands. This is not applicable in 
the exclusive use case. 
 
Figure A14-3. Apron capacity estimation tool - output for hub airports 
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
181 
 
In order to obtain capacity estimates for different airport types with respect to the nature 
of traffic at the airport, it is necessary to define properly stand occupancy times: 
1. For O/D airports the turnaround times matrix should be filled in with different 
values depending on the user and aircraft class. The appropriate separation time 
depending on the user should be entered in the separation time’s matrix. 
2. For HUB airport fields for coordinated flights in the turnaround times matrix are 
automatically filled in by ARRtw+MCT (derived from the maximum number of 
aircraft is calculated), while fields for other flights should be filled in with 
different values depending on the user and aircraft class. The appropriate 
separation time should be entered in the separation time’s matrix.  
When calculating apron capacity the tool recognizes when to use WRC (filled in the 
hubbing parameters input set), and when SOT (determined as ARRtw+MCT+ST), 
depending on the case modeled (exclusive or preferential).  
The tool delivers apron hourly saturation capacity, which is “divided” between users 
accounting for their share in total demand. For hub airports, the tool also provides apron 
capacity per WRC, which gives a better picture of the system performance than hourly 
capacity, in this case.  
 
  
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
182 
 
Appendix 15 – Stand allocation at Zurich Airport after Switzerland entry into the 
Schengen zone54 
 
Figure A15-1. Schengen user concept at Zurich Airport, Switzerland - final concept plan (ZRH, 2008) 
 
Figure A15-2. Schengen user concept at Zurich Airport, Switzerland - transition period (ZRH, 2008) 
                                                 
54 Provided through the courtesy of Operations Planning, Planning and Engineering, Zurich Airport 
pier E pier A pier B busgates A busgates B busgates D
Non-
Schengen
Schengen
A01-10
Swiss
long-haul
1 Swiss
long-haul
2
Swiss
short- /medium
1 Swiss
short- /medium
2
Swiss
mix-rotations
Star Alliance
long-haul
var. airlines
long-haul
Star Alliance
short- /medium
1
Star Alliance
short- /medium
3
var. airlines
short- /medium-
long-haul
B01-10
B22-29
var. airlines
short- /medium
A50
overflow pier
and busgates
Star Alliance
short- /medium
2
Star Alliance
short- /medium
2
Star Alliance
short- /medium
1
var. airlines
short- /medium
var. Airlines
short- /medium-
long-haul
Mix-
Rotations
Swiss
short- /medium
3
Swiss
short- /medium.
1
Star Alliance
short- /medium
3
Swiss
short- /medium
2
Swiss
short- /medium
3
var. airlines
mix-rotations
var. airlines
mix-rotations
overflow pier
and busgates
var. airlines
short- /medium
B22-29
pier E pier A busgates A busgates B
Non-
Schengen
Schengen
A01-10
Swiss
long-haul
Swiss
short- /medium
1
Star Alliance
long-haul
var. airlines
long-haul
Star Alliance
short- /medium
1
Star Alliance
short- /medium
2
B01-10
A50
overflow pier A
and Busgate B
Star Alliance
short- /medium
1
Mix-
Rotations
Swiss
short- /medium.
2
Swiss
short- /medium
1
Star Alliance
short- /medium
2
Swiss
Mix-Rot.
var. airlines
short- /medium
Swiss
short- /medium
2
var. airlines
short- /medium
overflow
long-haul
var. airlines
mix-rotations
var. airlines
short- /medium
1var. Airlines
short- /medium
2
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
183 
 
 
Figure A15-3. Schengen user concept at Zurich Airport, Switzerland - final concept (ZRH, 2012b) 
 
   
Airport airside balanced capacity usage and planning  Bojana Mirkovic 
184 
 
12 Biography 
Bojana Mirković was born in G. Milanovac, 28.07.1980. She finished primary and 
secondary school in Ljig. 
She graduated from Faculty of Transport and Traffic Engineering in 2005 and obtained 
her Master degree in 2008, both from the Department of Air Transport and Traffic. The 
research topic for her PhD was accepted by the University of Belgrade in December 
2010. 
She has been employed at Department of Air Transport and Traffic since December 
2005, as a teaching and research assistant. During eight years of her teaching experience 
she has been engaged in labs for all undergraduate courses at the Division of Airports 
Traffic Safety and the Airports 3 course of the Master’s studies.  
She has interned at three European airports: Tenerife Sur, Spain (2004), Munich 
Airport, Germany (2005) and Zurich Airport, Switzerland (2008); and attended ten 
certified aviation courses/seminars/workshops, one of which she has organized.  
She has taken part in 12 science and professional projects.  
She is the coauthor of two textbooks, one practicum and 12 published and/or presented 
papers.  
Her fields of interest are airports - design, planning and operations; and air traffic 
analysis, modeling and forecasting.  
 
 
lzjava o autorstvu
Potpisana:
Broj upisa:
Boiana Mirkovic
./
lzjavljujem
da je doktorska disertacija pod naslovom
BALANSTRANo ront5ceNJE KApAcTTETA I pLANTRANJE RAzvoJA
ELEMENATA AERODROMA
o
o
o
o
re^J ltat sopstvenog istraZivackog rada.
da predlodena disertacija u celini ni u delovima nije
bilo koje diploma prema studijskim programima drugih
da su rezultati korektno navedeni i
da nisam kr5ila autorska prava i koristila intelektualnu
bila predloZena za dobijanje
visoko5kolskih ustanova.
svojinu drugih lica.
Potpis doktoranda
U Beogradu, 42.4[,2o1b,
l4ava o istovetnosti Stampane i elektronske verzije
doktorskog rada
f me i prezime autora: Boiana Mirkovic
Broj upisa: ,/
Studijski program'. ./
Naslov rada: Balansirano koriS6enie kapaciteta i planiranie razvoia
elemenata aerodroma
dr Voiin To5icMentor:
Potpisana: Boian? Mirkovic
izjavljujem da je Stampana verzija mog doktorskog rada istovetna elektronskoj verziji koju
sam predala za objavljivanje na portalu Digitalnog repozitorijuma Univerziteta u
Beogradu.
Dozvoljavam da se objave moji lidni podaci vezani za dobijanje akademskog naziva
doktora nauka, kao 5to su ime i prezime, godina i mesto rodenja idatum odbrane rada.
Ovi lidni podaci mogu se objaviti na mreZnim stranicama digitalne biblioteke, u
elektronskom katalogu iu publikacijama Univerziteta u Beogradu.
Potpis doktoranda
u BeogradLr, 42. {2,9.oib.
l4ava o kori5cenju
Ovla56ujem Univezitetsku biblioteku ,,Svetozar Markovid" da u Digitalni repozitorijum
Univerziteta u Beogradu unese moju doktorsku disertaciju pod naslovom:
BALANSIRANO KORISCENJE KAPACITETA I PLANIRANJE RAZVOJA
ELEMENATA AERODROMA
koja je moje autorsko delo.
Disertaciju sa svim prilozima predala sam u elektronskom format pogodnom za trajno
arhiviranje.
Moju doktorsku disertaciju pohranjenu u Digitalni repozitorijum Univeziteta u Beogradu
mogu da koriste svi koji po5tuju odredbe sadrZane u odabranom tipu licence Kreativne
zajednice (Creative Commons) za koju sam se odludila.
1. Autorstvo
2. Autorstvo - nekomercijalno
3. Autorstvo - nekomercijalno 
-bez prerade
(rAutorstvo - nekomercijalno - deliti pod istim uslovima
5. Autorstvo 
- 
bez prerade
6. Autorstvo 
- 
deliti pod istim uslovima
Potpis doktoranda
U Beogradu, 42,4L.2o4b.