A comparative study of GA and PSO approach for cost optimisation in product recovery systems
研究比较了遗传算法和粒子群优化在解决产品回收系统成本最小化问题上的效果，通过混合整数线性规划模型和快消品行业案例验证了成本降低。
A product recovery system is proposed to reduce the bulk of waste sent to landfills by retrieving materials and parts of obsolete products for using them in remanufacturing and recycling. Product recovery is a significant strategy for enhancing customer satisfaction with regard to environmental concerns. Considering the fact that some products are returned, it becomes challenging to analyse whether to manufacture a new product or to rework the returned product at every step of the product recovery chain. Our approach uses a mixed integer linear programming model with the genetic algorithm and particle swarm optimisation, where two meta-heuristic algorithms are introduced for solving the MILP problem. Here, a recovery scenario is modelled, subject to the time and type of product to be processed. The study is intended to enhance the overall productivity of the product recovery chain. To demonstrate the approach, a case study is presented in the fast-moving consumer goods industry in which the proposed model demonstrates a reduction in the overall cost in the product recovery chain.