A Metaheuristic-Based Approach for the Capacitated Supply Chain Network Design Problem Including Imperfect Quality and Rework

Abstract

A growing number of organizations have realized the importance of quantifying costs associated to product quality and optimizing their supply chains based not only on operational and logistics costs but also on quality-related costs or Cost of Quality (COQ). This paper presents a novel capacitated Supply Chain Network Design (SCND) model, known as the SCND-COQ model, which quantifies the overall economic profit of the supply chain while accounting for quality-related costs. Quality-related costs are computed from analytical expressions for tracking the Supply Chain (SC) quality level and quantifying the prevention, inspection, rework, failure and opportunity costs, which, in turn, depend on the internal operational decisions within the manufacturing plants. Two metaheuristic solution procedures, based on the Simulated Annealing (SA) and the Genetic Algorithm (GA), with calls to a nonlinear solver are proposed for identifying near-optimal SCNDs since maximizing the output of the SCNDCOQ model can be classified as an NPO-complete problem. The effectiveness of the proposed solution procedures is demonstrated through comprehensive numerical experiments. Based on computational results, the GA-based procedure outperformed the SA-based procedure for all the tested instances in terms of solution quality. The results show that quality-related costs may account up to 11% of the overall profit and the selected business entities differ when including quality-related costs in the SCND decision-making process.