Cross-trained versus specialized agents in an inbound call centre: a simulation-based methodology for trade-off analysis

While advances in call-centre technologies offer increased capabilities and efficiencies, the expanded possibilities for alternative configurations and operating policies complicate call-centre design and management correspondingly. Discrete-event simulation can capture these complexities and assist call-centre management in leveraging new technologies. In this study, we describe a simulation-based methodology developed for the inbound call centres of a leading catalog retailer of consumer electronics. Management sought to understand the implications of exploiting the capabilities of advanced telephone switches to migrate from a staff comprised entirely of cross-trained agents to a staff of dedicated product specialists. At issue was the trade-off between the efficiency of the single-channel generalist configuration and the potential effectiveness of a dual-channel specialist configuration. We describe the resolution of a range of technical issues embodied in this methodology, including modelling non-stationary call-arrival processes, deriving service-delay distributions from aggregate call-duration data, modelling call flow and control processes, adaptation of a framework for proper statistical comparison of multiple alternatives, and a software utility developed to integrate experiment design, run control, and output analysis.