Parametric Forecasting and Stochastic Programming Models for Call-Center Workforce Scheduling
Abstract
We develop and test an integrated forecasting and stochastic programming approach to workforce management in call centers. We first demonstrate that parametric forecasts, discretized using Gaussian quadrature, can be used to drive stochastic programs whose results are stable with relatively small numbers of scenarios. We then extend our approach to include forecast updates and two-stage stochastic programs with recourse and provide a general modeling framework for which recent, related models are special cases. In our formulations, the inclusion of multiple arrival-rate scenarios allows call centers to meet long-run average quality-of-service targets, and the use of recourse actions helps them to lower long-run average costs. Experiments with two large sets of call-center data highlight the complementary nature of these elements.
