考虑不确定生产因素的表面组装生产线负载平衡优化模型

Load balance optimization model of surface-mount production line considering uncertain production factors

中文关键词  负载平衡  蒙特卡洛法  神经网络  遗传算法

英文关键词  load balancing  Monte Carlo method  neural networks  genetic algorithm

中文摘要

      针对表面组装生产中的不确定因素造成企业订单完成时间滞后问题, 本文设计并实现了一种考虑不确定 生产因素的生产线负载平衡优化模型. 首先, 以不确定生产因素的历史样本数据作为随机模拟样本预估出不确定 生产因素对订单完成造成的滞后时间; 其次, 优化元器件贴装工位分配方案并以任务完成作为触发事件模拟生产 线实际运行得到动态生产计划; 再次, 根据动态生产计划计算出模型适应度值后, 采用遗传算法对模型适应度值进 行启发式寻优获得最优动态生产方案. 最后, 利用表面组装生产线试例对该模型进行验证, 结果表明, 该模型可准确 预测产线各时段的生产任务、任务量及各器件贴装工位, 有效提高了企业生产效率.

英文摘要

      Aiming at the problem that uncertain production factors cause a lag in order fulfillment for surface-mount manufacturers, a load balance optimization model of production line considering uncertain production factors is designed. First, the historical sample data of uncertain production factors are used as a random simulation sample to predict the lag time of order completion caused by uncertain production factors. Second, after optimizing the component placement workstation allocation scheme, the task completion is used as the trigger event to simulate the actual operation of the production line to obtain a dynamic production plan. Third, after calculating the model fitness value according to the dynamic production plan, the genetic algorithm is used to heuristically optimize the model fitness value to obtain the optimal dynamic production plan. Finally, the model is validated by using a surface-mount production line test case. The result shows that the model can accurately predict the production tasks, task volume and placement stations of components in each period of the production line and can effectively improve the production efficiency of enterprises.