Robust receding horizon control strategy for replenishment planning of pharmacy robotic dispensing systems

Abstract

This paper presents a robust receding horizon control strategy (RHC-RO) to enhance replenishment planning and inventory control of robotic dispensing systems in central fill pharmacies (CFPs). Replenishment in CFPs is a key process greatly influenced by several stochastic factors, such as demand volume and process times. In this research, a robust mixed integer quadratic programming (RMIQP) model is proposed to determine the number of allocated canisters and the schedule of replenishment operations considering multiple scenarios. A receding horizon control (RHC) mechanism, which divides the optimization horizon into smaller time windows, is applied to enhance the solution quality and reduce the computational burden. The proposed RHC-RO strategy is evaluated using simulation against offline, robust optimization (RO), and RHC strategies in terms of total replenishment costs. The results indicate that RHC-RO outperforms the offline, RO, and RHC strategies by generating 19.7%, 18.3%, and 5.6% less replenishment costs on average, respectively. The results also show that the RHC-RO strategy enables timely, accurate, and robust replenishment decisions.