基于改进遗传算法的双边装配线平衡

Two-sided assembly line balancing based on modified genetic algorithm

针对目前研究较少的双边装配线平衡问题,研究了双边装配线的特点及其对平衡的特殊要求,建立了双边装配线平衡问题的数学模型。根据双边装配中任务具有操作方位约束,以及工位上分配任务的操作顺序与平衡结果直接相关等特点,提出了相应的符合问题特性的遗传算法。该算法采用基于序列、任务及其分配方位组合的编码方法,运用可行的交叉与变异算子,使搜索过程仅在可行解空间内进行,提高了效率。算例结果验证了算法的有效性。

Aiming at the lack of research on two-- sided assembly line balancing problem status, the characteristics of two--sided assembly line were analyzed and the special requirements of balancing were studied, the corresponding mathematical model was built. A genetic algorithm was proposed according to the features in two--sided assembly, such as the tasks having operational direction constraints and the operating sequence of tasks in stations correlated with the balancing result. A combinational coding scheme, which was based on sequence and consisted of the task and the assigned direction of the task, was adopted. The feasible genetic operators, which only explored in the feasi- ble solution space, were adopted. Experiments were carried out to demonstrate the performance of the proposed method.