An Improved Fruit Fly Optimization Algorithm for Disassembly Lines Requiring Multiskilled Workers

Waste recycling is an important part of resource reuse and environmental protection. The study of disassembly lines deals with the process of recycling and remanufacturing end-of-life products. The performance of a disassembly line is affected by many factors, especially the operation cost of workstations, the precedence relationships among disassembly tasks, the skill level of workers, and their learning speed. This study considers the learning effect of disassembly workers, establish a mixed integer programming model of the disassembly balancing problem, and explores the search for optimal solution. It allocates tasks and multiskilled workers on workstations to maximize disassembly profits in the disassembly process. To solve it, an improved fruit fly optimization algorithm is proposed, and three methods are designed for the smell search. At the same time, the visual search is also designed to avoid the problem of falling into local optimum. The validity and effectiveness of the proposed algorithm are verified with experiments that compare the results with CPLEX, a well-known IBM optimizer, and some popular peer algorithms.