Abstract:
Accurate forecasting of call arrivals is critical for staffing and scheduling of a telephone call center. We develop methods for interday and dynamic intraday forecasting of incoming call volumes. Our approach is to treat the intraday call volume profiles as a high-dimensional vector time series. We propose to first reduce the dimensionality by singular value decomposition of the matrix of historical intraday profiles and then apply time series and regression techniques. Both interday (or day-to-day) dynamics and intraday (or within-day) patterns of call arrivals are taken into account by our approach. Distributional forecasts are also developed. The proposed methods are data-driven, and appear to be robust against model assumptions in our simulation studies. They are shown to be very competitive against existing approaches in out-of-sample forecast comparisons using real data sets. Our methods are computationally fast and therefore it is feasible to use them for real-time dynamic forecasting.