Show simple item record

Simulation models based on agents as decision making support in electronic commmerce

dc.contributor.advisorMarković, Aleksandar
dc.contributor.otherDespotović-Zrakić, Marijana
dc.contributor.otherStanojević, Milorad
dc.creatorČavoški, Sava
dc.date.accessioned2017-12-12T15:45:31Z
dc.date.available2017-12-12T15:45:31Z
dc.date.issued2017-03-30
dc.identifier.urihttp://eteze.bg.ac.rs/application/showtheses?thesesId=5429
dc.identifier.urihttps://fedorabg.bg.ac.rs/fedora/get/o:16808/bdef:Content/download
dc.identifier.urihttp://vbs.rs/scripts/cobiss?command=DISPLAY&base=70036&RID=515608986
dc.identifier.urihttp://nardus.mpn.gov.rs/123456789/8948
dc.description.abstractPredmet istraživanja u ovom radu su simulacioni modeli zasnovani na agentima (ABMS) i njihova primena u oblasti elektronskog poslovanja. Oblast elektronske trgovine, zbog svoje dinamičnosti i stalnog razvoja, zahteva primenu brojnih alata za podršku odlučivanju. Ovakvi alati moraju biti u mogućnosti da u kratkom vremenskom periodu obrade velike količine podataka, koje generišu e-commerce sistemi i omoguće da se dobijeni rezultati pretoče u kvalitetne poslovne odluke. Metodologija simulacije zasnovana na agentima može da pruži značajan doprinos brzini i kvalitetu donošenja poslovnih odluka u elektronskom poslovanju. Modeli razvijeni za potrebe ovog rada imaju za cilj da naprave pomak u korišćenju praktičnih alata za evaluaciju sistema B2C online prodaje, tako što omogućavaju razmatranje posledica različitih politika u upravljanju e-commerce sajtom na ponašanje potrošača, posećenost sajta, obim prodaje, ostvarene prihode i sl. Povezivanjem oblasti modelovanja baziranog na agentima i elektronske trgovine, otvaraju se nove mogućnosti za kvalitetnije sagledavanje ponašanja i uzroka ponašanja u e-commerce sistemima. Jedan od ciljeva ovog rada jeste da istraži kako različite navike potrošača u donošenju odluka o kupovini utiču na kompleksnost njihovih navika prilikom kupovine na Internetu. Razvoj modela elektronske trgovine je duži niz godina predmet brojnih istraživanja. U naučnoj literaturi se često susreće primena regresione analize, kao jedan od najčešćih pristupa za sagledavanje uticaja ključnih faktora na uspešnost odabranog modela elektronske trgovine. U svom radu A Dynamic Resource Allocation on Service Supply Chain Kim, Choi 2001 prikazali su primenu metodologije dinamike sistema na modelima elektronske trgovine. Da bi se unapredila postojeća rešenja i istražile nove mogućnosti u cilju podrške boljim poslovnim odlukama, poslednjih godina u istraživanjima je sve prisutnija primena modela zasnovanih na agentima u analizi e-commerc poslovnih modela. U svom radu iz 2009 Agent-Based and Individual-Based Modeling, Grimm i Railsback pokazali su da agenti mogu da adaptiraju svoje ponašanje u odnosu na trenutno stanje okruženja i druge agenate. Grimm, Berger i drugi autori u radu A standard protocol for describing individual-based and agent-based models, iz 2006 godine pokazali su da je jedna od najbitnijih karakteristika agenata njihovo adaptivno ponašanje. Prema tome, složena i dinamična okruženja, kao što su on-line tržišta, mogu se uspešno modelovati i simulirati ovom metodologijom...sr
dc.description.abstractThis paper analyses simulation models based on agents and their application in field of e-commerce. The continuous development and dynamics in the field of e-commerce requires application of advanced decision-making tools. These tools must be able to process, in a short time period, a large amount of data generated by the e-commerce systems and enable the use of acquired data for making qualitative business decisions. Models developed for this paper aim to make a change in using practical tools for the evaluation of the B2C online sale systems by allowing the evaluation of the consequences of consumer behaviour and different policies in management of e-commerce sites, frequency of site visits, scope of sale and achieved income. One of the aims of the paper is to examine how consumer habits influence the complexity of their habits in buying online. The development of the electronic commerce model has long been a subject of numerous research endeavours. The scientific literature often uses regression analysis as one of the most common approaches in recognizing the impact of key factors upon the success of a selected model of electronic commerce. The application of methodology system dynamics in modelling in the field of e-commerce may also be found in the paper “A Dynamic Resource Allocation on Service Supply Chain”, co-authored by Kim and Choi. In order to improve the existing approach in analysis of e-commerce of business systems, the application of agent based models is becoming more prominent in recent years. In their work from 2009 “Agent-Based and Individual-Based Modelling” Grimm and Railsback proved that agents may adjust their behaviour depending on the situation in the environment and the behaviours of other agents. Grimm, Berger and others in 2006 in their A standard protocol for describing individual-based and agent-based models demonstrated that one of the most important features is the ability to adjust. Thus, a complex and dynamic environment, such as online markets may be successfully modelled and simulated by using this methodology. One of the most famous models was applied in practice and developed by North and Macal for the purposes of Procter & Gamble (Multi-Scale Consumer Market Model). In their 2009 paper “Agent-based simulation of consumer purchase decision-making and the decoy effect”, Tao and David Zhang used the agent-based simulation model to present the effect of introducing a new product on the market to serve as a decoy...en
dc.formatapplication/pdf
dc.languagesr
dc.publisherУниверзитет у Београду, Факултет организационих наукаsr
dc.rightsAutorstvo-Nekomercijalno-Bez prerade 3.0 Srbija (CC BY-NC-ND 3.0)
dc.sourceУниверзитет у Београдуsr
dc.subjectsimulacioni modeli zasnovani na agentimasr
dc.subjectsimulation models based on agentsen
dc.subjectB2Cen
dc.subjectelectronic commerceen
dc.subjectcustomer behavioren
dc.subjectB2Csr
dc.subjectelektronsko poslovanjesr
dc.subjectponašanje potrošačasr
dc.titleSimulacioni modeli zasnovani na agentima kao podrška odlučivanju u elektronskom poslovanjusr
dc.title.alternativeSimulation models based on agents as decision making support in electronic commmerceen
dc.typePhD thesis
dcterms.abstractМарковић, Aлександар; Деспотовић-Зракић, Маријана; Станојевић, Милорад; Чавошки, Сава; Симулациони модели засновани на агентима као подршка одлучивању у електронском пословању; Симулациони модели засновани на агентима као подршка одлучивању у електронском пословању;


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record