Staffing a service system with appointment-based customer arrivals

Appointment systems are widely used to facilitate customers’ access to services in many industries such as healthcare. A number of studies have taken a queueing approach to analyse service systems and facilitate managerial decisions on staffing requirements by assuming independent and stationary customer arrivals. This paper is motivated by the observation that the queueing‐based method shows relatively poor performance when customers arrive according to their appointment times. Because customer arrivals are dependent on their appointment times, this study, unlike queueing‐based methods, conducts a detailed analysis of appointment‐based customer arrivals instead of making steady‐state assumptions. We develop a new model that captures the characteristics of appointment‐based customer arrivals and computes the probability of transient system states. Through the use of this model, which relaxes stationary and independent assumptions, we propose a heuristic algorithm that determines staffing requirements with aims to minimizing staff‐hours while satisfying a target service level. The simulation results show that the proposed method outperforms the queueing‐based method.