A particle swarm approach for optimizing a multi-stage closed loop supply chain for the solar cell industry. (February 2017)

• Record Type: Journal Article
• Authors: Chen, Yi-Wen; Wang, Li-Chih; Wang, Allen; Chen, Tzu-Li
• Is Part Of: Robotics and computer-integrated manufacturing. Volume 43(2017)
• Page Start: 111
• Page End: 123
• Publication Date: 2017-02
• Subjects: Closed-loop supply chain design -- Multi-objective searching -- Particle swarm optimization -- Solar energy industry; Robots, Industrial -- Periodicals; Computer integrated manufacturing systems -- Periodicals; Robotics -- Periodicals; Robots industriels -- Périodiques; Productique -- Périodiques; Robotique -- Périodiques; 670.285
• DOI: 10.1016/j.rcim.2015.10.006
• Languages: English
• ISSNs: 0736-5845

Abstract

Abstract: In order to implement sustainable strategies in a supply chain, enterprises should provide highly favorable and effective solutions for reducing carbon dioxide emissions, which brings out the issues of designing and managing a closed-loop supply chain (CLSC). This paper studies an integrated CLSC network design problem with cost and environmental concerns in the solar energy industry from sustainability perspectives. A multi-objective closed-loop supply chain design (MCSCD) model has been proposed, in consideration of many practical characteristics including flow conservation at each production/recycling unit of forward/reverse logistics (FL/RL), capacity expansion, and recycled components. A deterministic multi-objective mixed integer linear programming (MILP) model capturing the tradeoffs between the total cost and total CO2 emissions was developed to address the multistage CSLC design problem. Subsequently, a multi-objective PSO (MOPSO) algorithm with crowding distance-based nondominated sorting approach is developed to search the near-optimal solution of the MCSCD model. The computational study shows that the proposed MOPSO algorithm is suitable and effective for solving large-scale complicated CLSC structure than the conventional branch-and-bound optimization approach. Analysis results show that an enterprise needs to apply an adequate recycling strategy or energy saving technology to achieve a better economic effectiveness if the carbon emission regulation is applied. Consequently, the Pareto optimal solution obtained from MOPSO algorithm may give the superior suggestions of CLSC design, such as factory location options, capacity expansion, technology selection, purchasing, and order fulfillment decisions in practice. Highlights: An integrated CLSC problem by using PSO algorithm in the solar energy industry has been studied. A deterministic multi-objective mixed integer programming model has been established. The results Pareto CLSC solutions between CPLEX and MOPSO have been compared. MOPSO provides a rapid algorithm to search Pareto solution compare to CPLEX in large scale issues.