Abstract: For the multiple assembly line system composed by main line and feeder line,the influence of feeder line's cycle time and buffer capacity on output of main line was studied,and a method of optimizing feeder line was proposed.By regarding dynamic cycle time constraints and positional constraints,a mathematical model was established to minimize the buffer between main line and feeder line and the buffer between workstations of feeder line.A two-step-decoding genetic algorithm was used to solve the model.To make the results more realistic,a method of segmental coding and segmental decoding was designed for positional constraints.The experimental results showed that the proposed method could not only improve the performance of feeder line,but also improve the output of main line.

Key words: feeder line, positional constraints, cycle time constraints, buffer capacity optimization, genetic algorithms

摘要： 针对由主装线和分装线构成的多装配线系统,研究了分装线节拍及缓冲区容量对主装线产能的影响,提出优化分装线的方法。该方法以动态变化的节拍作为分装线的节拍约束,同时考虑操作具有位置约束,以线间缓冲区容量与线上缓冲区总容量之和最小为目标建立数学模型,并设计两步解码的遗传算法对模型进行求解。为了使优化结果更加符合工程实际,采用分段编码和分段解码的方式求解操作位置固定的问题。实例结果表明,该方法不但提升了分装线的性能,而且有效提高了主装线的产能。

关键词: 分装线, 位置约束, 节拍约束, 缓冲区容量优化, 遗传算法