A fuzzy inference based scenario building in two-stage optimization framework for sustainable recycling supply chain redesign. (1st March 2021)

• Record Type: Journal Article
• Authors: Feitó-Cespón, Michael; Costa, Yasel; Pishvaee, Mir Saman; Cespón-Castro, Roberto
• Is Part Of: Expert systems with applications. Volume 165(2021)
• Publication Date: 2021-03-01
• Subjects: Fuzzy inference system -- Reverse supply chain design -- Plastic recycling -- Scenario generation; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2020.113906
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: A two-stage optimization framework is proposed for reverse supply chain design. We introduce the Fuzzy Inference System for scenario generation. We propose the solution robustness analysis based on Customer Service Level concept. A redesign problem of plastic recycling supply chain is studied. A realistic case study in Cuba is conducted. Abstract: This paper aims to develop a scenario-based optimization framework to deals with several issues related to redesign sustainable reverse supply chain; focused particularly in the epistemic uncertainty of supply and demand input parameters and their relationship with supply chain performance, both important for the redesign problem. The two-step optimization framework starts with a Fuzzy Inference System methodology for scenario generation that faces the lack of information, and the necessity of estimate the expected operation cost and environmental impact. Then we use generated scenarios into the epsilon-constraint method which solves a multi-objective model to obtain a relevant set of solutions. After solving the second optimization model, we propose to analyze the robustness of the achieved redesign solutions considering a customer satisfaction approach. The computational experiments show that our proposed framework supports better the inclusion of scenarios for redesigning the plastic recycling supply chain. Furthermore, we study a real-life plastic recycling problem in Cuba which demonstrates that the framework is able to support the redesign decision making with robust solutions sensitive to the changes of the studied uncertain parameters.