Surgery scheduling in outpatient procedure centre with re-entrant patient flow and fuzzy service times

Abstract

To meet the increase in demand for outpatient surgical services, surgery scheduling for outpatient procedure centres (OPCs) has recently attracted considerable attention in both the healthcare industry and the academic community. This paper considers a novel OPC daily surgery scheduling problem (ODSSP) to minimise the average recovery completion time of all patients. To satisfy the OPC surgical practice, patient intake and recovery are applied in the same area for more resource flexibility, and uncertain service times for intake, surgical procedures and recovery are considered. Owing to the similarities shared between healthcare delivery systems and production systems, ODSSP is formulated as a two-stage no-wait re-entrant hybrid flow shop scheduling problem with fuzzy service times. Considering the NP-hardness of such scheduling problem, a new hybrid meta-heuristic (GA-BAVNS) is employed to obtain detailed daily OPC surgery schedules. To achieve greater balance between exploration and exploitation in the search space, GA-BAVNS hybridises a genetic algorithm (GA) and a variable neighbourhood search (VNS) to schedule outpatients for surgical services. To improve the local search performance of the VNS, six novel block-based neighbourhood structures are employed to generate neighbourhood solutions. Moreover, an adaptive neighbourhood change procedure (ANCP) is employed to systematically change the search order of neighbourhood structures for better solutions. Computational results on a set of test problems indicate the superiority of the proposed GA-BAVNS.