The impact of prolonged service time on flexibility configurations

Abstract

Problem definition: Resource pooling is often introduced in service systems to cope with the variability in customer demand, given a fixed service capacity for each service provider. The initial motivation for a more flexible system is to use the resources efficiently, that is, assign customers whose dedicated resources are fully occupied to available non-dedicated resources (referred to as off-service placement). In such a setting, the service system manager expects more customers to be served during a fixed period. However, recent empirical observations find that the service time at the non-dedicated service provider is significantly prolonged compared with that at the dedicated service provider. In this study, we construct a two-server stochastic model to assess the impact of prolonged service time and other parameters, such as system workload and asymmetry level, on flexibility configurations. Methodology/results: We derive conditions that describe the optimal flexibility configuration within the space of system parameters. Our results indicate that a partially flexible system might outperform a fully flexible one, especially in an asymmetrical setup with prolonged service times. Managerial implications: These insights from the two-server model provide more efficient flexibility design in complex multi-class, multi-server service systems.

Keywords: Resource pooling, off-service placement, prolonged service time, flexibility designs, queueing models