Accurate estimation of times between arrivals is a key aspect of process simulations and thus efficient decision making. Times between arrivals typically exhibit strong autocorrelation structure which is commonly ignored in the queueing theory. In order to capture the time dependence accurately, we utilize a Generalized Autoregressive Score (GAS) model based on the generalized gamma distribution. Once the process of arrivals is estimated, a process assessment can be performed using process simulations. The results from an empirical study of an online bookshop in Prague, Czech Republic pointed out insufficient resources allocated for the pre-processing and the final stage.
How to Cite
Arrival Process, Duration Model, Generalized Autoregressive Score Model, Process Simulation
Creal, D., Koopman, S.J. & Lucas, A. (2013). Generalized autoregressive score models with applications. Journal of Applied Econometrics, 28(5), 777-795.
Harvey, A.C. (2013). Dynamic models for volatility and heavy tails: with applications to financial and economic time series (Vol. 52). Cambridge University Press.
Tomanová, P. (2019). Clustering of Arrivals and Its Impact on Process Simulation. The 15th International Symposium on Operational Research SOR'19, Proceedings. Ljubljana: Slovenian Society Informatika, Section for Operational Research, 314-319.