Simulation of a telephone operator call center

Optimizacija poslovnih procesov je v današnjem času pravzaprav stalnica. Nobena izjema niso klicni centri, v katerih želijo z minimalnimi resursi zagotoviti zadovoljstvo strank s storitvami. V diplomski nalogi smo skušali ugotoviti optimalno število operaterjev v klicnem centru, s katerimi bi zagotovili, da stranke na pogovor ne bi čakale več kot 20 sekund in bi bile vse postrežene. S pomočjo programa AnyLogic smo razvili simulacijski model klicnega centra, ki omogoča preverjanje različnih scenarijev v modelu. V analizo so vključena 4 različna časovna obdobja delovanja klicnega centra, uporabljeni pa so dejanski podatki, pridobljeni iz informacijskega sistema klicnega centra. Za primerjavo rezultatov smo najprej uporabili vhodne podatke o časih med prihodi in časih strežbe v obliki frekvenčne porazdelitve, nato pa še v obliki povprečij eksponentne porazdelitve. Rezultate simulacij smo primerjali tudi z rezultati predhodne raziskave, v katerih je bilo optimalno število operaterjev določeno s pomočjo modelov množične strežbe. Rezultati simulacije z vhodnimi podatki v obliki frekvenčnih porazdelitev so bistveno drugačni kot rezultati simulacije z vhodnimi podatki v obliki povprečij eksponentne porazdelitve. Vendar pa so rezultati simulacije z vhodnimi podatki v obliki povprečij eksponentne porazdelitve skoraj identični rezultatom iz modelov množične strežbe. Razloge za razlike oziroma primernost modelov za napovedovanje števila zaposlenih operaterjev bi bilo smiselno identificirati na osnovi primerjave z realnim stanjem v klicnem centru.Optimization of business processes is nowadays essentially a constant. Call centers that want to ensure customer satisfaction with their services with minimal resources are no exception. In this thesis we tried to determine the optimum number of operators in a call center, in order to ensure that the parties would not wait more than 20 seconds for the conversation and that they would all be served. Using the program AnyLogic we developed a simulation model of the call center, which allows to check various scenarios in the model. The analysis includes 4 different time periods of the call center operation, and uses the actual data obtained from the call center\u27s information system. To compare the results, we first used the input data on the times between arrivals and about the times of serving in the form of the frequency distribution, and then in the form of averages of the exponential distribution. We compared the results of the simulations with the results of previous research, in which the optimal number of operators was determined by using the models of queuing. The simulation results by using the input data in the form of frequency distributions are materially different than the simulation results by using the input data in the form of averages of the exponential distribution. However, the simulation results by using the input data in the form of averages of the exponential distribution are almost identical to the results of the queuing models. It would be reasonable to identify the reasons for the differences or for the suitability of the models for predicting the number of operators by comparing with the real situation in the call center