A Network Design Approach for Circular Medical Waste Management using Two-stage Stochastic MILP Optimization

Abstract

The escalating production of medical waste, attributed to increased consumption levels, changing lifestyles, and natural disasters, poses a threat to the environment and human health. Among the hazardous waste categories, medical waste from healthcare centers stands out as a significant concern, demanding effective management strategies. This research addresses the challenges of fluctuating and uncertain waste generation, diverse waste types, incompatible handling practices, container and truck management, high costs, and the imperative for a circular waste management approach in the Medical Waste Management System (MWMS). A two-stage Stochastic Mixed-Integer Linear Programming (MILP) model is proposed to optimize the network while considering revenue generation from recycling, Waste-to-Energy (WTE) conversion, and reusing practices. The model's efficacy is demonstrated through a detailed case study implemented in Hamilton, Ontario, Canada. Data-driven parameter estimation, treatment technology selection, and revenue estimation contribute to the robustness of the model. The use of the Sample Average Approximation (SAA) technique efficiently handles the computational complexities of the two-stage stochastic MILP model, making it applicable to large-scale problems. Moreover, the research showcases the data analysis undertaken to estimate parameters for the case study, providing valuable insights for effective waste management strategies.