Design of a reverse logistics network for recyclable collection in Nova Scotia using compaction trailers

Abstract

ABSTRACTAs municipal and regional initiatives for the collection of recyclable products, such as plastic, metal, glass, paper, etc., are increasing across the globe, designing collection networks to minimize collection costs and reduce associated greenhouse gas emissions are becoming important. In this study, a two-phase mathematical decomposition model for the collection of multiple recyclable products using compaction trailers is presented. In the first phase of the decomposition, compaction trailer routes for each product type are generated by repeatedly solving instances of the route-generation subproblem. In the second phase, a mixed integer linear programming is formulated for route selection and solved to determine the optimal combination of scenarios generated in the first phase to minimize total collection costs. A case study was conducted in the Canadian province of Nova Scotia using this approach for the collection of clear polyethylene terephthalate and aluminium cans. Numerical results show that the proposed optimization methodology can reduce current collection costs by 53.7%. The route configuration corresponding to this solution uses compaction trailers to pick-up approximately 85.1% of the two recyclable products investigated in this study, with the remaining collected using conventional trailers.