Statistical Analysis of Call-center Operational Data: Forecasting Call Arrivals, and Analyzing Customer Patience and Agent Service

Call centers are modern service networks in which agents provide services to customers via telephones. They have become a primary contact point between service providers and their customers. Inbound call centers take calls that are initiated by customers. Examples include phone operations that support reservation and sales for hotels, airlines, and car rental companies, as well as service centers of retail banks and brokerage firms. Outbound call centers initiate calls to outside parties, for example, those in organizations such as telemarketers, bill collection agencies, and marketing research and polling companies. The increasing importance and complexity of call-center operations have rendered call centers into a fertile ground for academic research, both empirical and theoretical. Two survey papers [1,2] and an on-line bibliography [3] have well documented the current state of relevant research on call-center operations and directions for future research. This article focuses on statistical research about inbound call centers, in the areas of call arrival forecasting and analysis of customer patience and agent service times. There has been empirical research on call centers that is done outside the operations community and covers behavioral and psychological aspects [2] and human resource issues [4]. Brown et al. [5] described a simplified path that each call follows through a typical inbound call center. A customer calls the telephone number associated with