基于滚动时域的两阶段恢复预定装配顺序模型
首发时间:2020-11-18
摘要:面向汽车制造商涂装工艺后的车身需按照预定装配顺序进入总装线的要求, 研究了线性缓冲区模式下涂装工艺后的车身物理重排序问题, 采用了1-Q和Q-1填充与释放的两阶段优化策略, 填充阶段运用了动态时间窗口规则, 设计了模拟退火粒子群算法, 释放阶段重点考虑异常超车信息, 并结合升序释放规则设计了搜索算法. 算例结果和比较分析表明, 两阶段优化策略对于求解该问题是有效的, 结合动态时间窗口的填充算法能选择更合适的车道, 结合升序释放规则的动态搜索算法解决了深度优先遍历超时的缺点, 同时获得了较规则更优的装配顺序.
关键词: 两阶段优化;恢复预定顺序;有限的节拍时间;动态时间窗口; 搜索算法
For information in English, please click here
Two-stage restoration model of scheduled assembly sequence based on rolling time domain
Abstract:\justifying Facing the requirement that the car body after the painting process of the automobile manufacturer must enter the final assembly line in the predetermined assembly sequence, the physical reordering of the body after the painting process in the linear buffer mode is studied. The two-stage optimization strategy of 1-Q and Q-1 filling and release is adopted. The dynamic time window rule is used in the filling phase, and the simulated annealing particle swarm algorithm is designed. The release phase focuses on abnormal overtaking information, and the search algorithm is designed in combination with the ascending release rule. The results of calculation examples and comparative analysis show that the two-stage optimization strategy is effective for solving this problem. The filling algorithm combined with the dynamic window can select a more suitable lane. The dynamic search algorithm combined with the ascending release rule solves the shortcomings of the depth-first traversal timeout. At the same time, a better assembly sequence than the rules is obtained.
Keywords: two-stage optimization restore predetermined sequence limited takt time dynamic time window search algorithm
引用
No.****
动态公开评议
共计0人参与
勘误表
基于滚动时域的两阶段恢复预定装配顺序模型
评论
全部评论0/1000