A MIP Model for Rolling Horizon Surgery Scheduling

Abstract

Most surgery scheduling is done 1 day in advance. Caused by lack of overall planning, this scheduling scheme often results in unbalanced occupancy time of the operating rooms. So we put forward a rolling horizon mixed integer programming model for the scheduling. Rolling horizon scheduling refers to a scheduling scheme in which cyclic surgical requests are taken into account. Surgical requests are updated daily. The completed surgeries are eliminated, and new surgeries are added to the scheduling list. Considering day-to-day demand for surgery, we develop a non-rolling scheduling model (NRSM) and a rolling horizon scheduling model (RSM). By comparing the two, we find that the quality of surgery scheduling is significantly influenced by the variation in demand from day to day. A rolling horizon scheduling will enable a more flexible planning of the pool of surgeries that have not been scheduled into this main blocks, and hence minimize the idle time of operating rooms. The strategy of the RSM helps balance the occupancy time among operating rooms. Using surgical data from five departments of the West China Hospital (WCH), we generate surgical demands randomly to compare the NRSM and the RSM. The results show the operating rooms' average utilization rate using RSM is significantly higher than when applying NRSM.

Keywords: MIP; Rolling horizon scheduling; Surgery scheduling.