Forecasting Arrivals to a Telephone Call Center

Accurate forecasting of call arrivals is critical for staffing and scheduling of a telephone call center. We develop methods for day-to-day and dynamic within-day forecasting of incoming call volumes. Our approach is to treat the intra-day call volume profiles as a highdimensional vector time series. We propose to first reduce the dimensionality by singular value decomposition of the matrix of historical intra-day profiles and then apply time series and regression techniques. Both day-to-day dynamics and within-day patterns of call arrivals are taken into account by our approach. The proposed methods are applied to a real data set and show substantial improvements over existing industry standards in an out-of-sample forecast comparison. Our methods are computationally fast and therefore it is feasible to use them for real-time dynamic forecasting.