Our study aims to address three main issues in the area of call center modeling; within day and intra-week call arrival forecasting for staffing purposes, abandonment behavior of different call center customer profiles, and inference in measures of performance of the most commonly used call center queuing models. All three issues will be addressed from a Bayesian perspective implying that all uncertainties including those about model parameters will be described probabilistically. In our first essay we introduce a discrete time Bayesian state space model with Poisson measurements for intra-day call arrivals. We present the properties of our model and develop Bayesian inference. In so doing, we provide analytically tractable expressions for sequential updating for parameters, for smoothing and prediction of call arrivals and discuss how the model can be used for inter-weekly forecasts. In our second essay, we consider modeling abandonment behavior in call centers for different customer profiles. We present several time to event modeling strategies and develop Bayesian inference for posterior and predictive analysis. For the third essay we consider the following. Queuing models have been extensively used in call center analysis for obtaining performance measures and for developing staffing policies. However, almost all of this work have been from a pure probabilistic point of view and have not addressed issues of statistical inference. In this paper, we develop Bayesian analysis of call center queuing models by describing uncertainty about system primitives probabilistically. We consider models with both patient and impatient customers and discuss their further extensions.
展开▼