面向成本的双边装配线平衡和员工分配

Cost-Oriented Two-Sided Assembly Line Balancing and Worker Assignment Problem

员工工资是装配成本中的重要组成部分,如何实现工资支出最小化得到越来越多的关注。同时在实际生产中,员工技能的差异也导致对相同操作不同的加工效率,进一步导致了员工最优化分配问题。针对面向成本的双边装配线平衡和员工分配问题,构建了最小化员工成本的数学模型,同步考虑装配线负载均衡,形成多目标的优化模型。提出一种改进的重启模拟退火算法,设计了新的邻域结构和重启机制。在邻域结构中,设计了3种邻域结构;在重启机制中,根据聚集距离优先选择较离散的个体,以获得分布性更好的帕累托解集。最后对7组随机生成的实例进行求解,将重启模拟退火算法与快速非支配排序遗传算法进行对比,验证所提算法的高性能。

Wage cost is a major expenditure in the assembly line, and how to reduce the wage cost draw more attention,. Meanwhile, multi-skill workers operate the same tasks with different operation time, which results in the optimization of worker assignnvent. To solve the cast-oriented two-sided assembly line balancing and worker assignment problem, a new mathematical model is developed to reduce the wage cost. The balance of the workload is also considered in this model, which results in a multi-objective optimtzation model. An improved restarted simulated annealing algorithm with new local search and restart mechanism is developed to solve this multi-objective problem. Three neighbor search structures are propased in the local search phase and the solution with the largest crowding distance is selected in the restart mechanism to replace the incumbent one to obtain well-spread Pareto-optimal set. Finally, the restarted simulated annealing algorithm is compared with the fast elitist non-dominated sorting genetic algorithm on seven set of problems, and experimental results demonstrate the effectiveness of the proposed method.