考虑学习效应的单人作业车间调度算法

单人负责多台机器的单一工序作业车间场景中,工人由于重复操作机器而产生学习效应.针对考虑依赖工件位置学习效应的单人单工序作业车间最小化最大完工时间的调度问题,建立一种混合整数规划模型.为解决该问题,设计一个考虑学习效应的贪婪算子,利用该算子构造两种贪婪算法,并提出一种基于贪婪的模拟退火算法.为衡量混合整数规划模型、贪婪算法和基于贪婪的模拟退火算法的性能,设计两种规模问题的数据实验.通过实验得出:现代混合整数规划模型求解器可以解决机器数量和工件总数量乘积小于75的小规模问题;基于贪婪的模拟退火算法求解此问题具有有效性,适用于各种规模的问题;间隔插入贪婪算法解决此问题速度较快,效果良好,可以应用于需要快速求解的场景.

Worker job shop scheduling algorithm considering learning effect

In some job shops where one worker needs to operate more than one machine, there is learning effect because the worker operates the machines repeatedly. For this one worker and one process production job shop scheduling problem minimizing makespan considering position-based learning effect, a mixed integer programming model is proposed. In order to solve this problem, a greedy operator considering learning effect is designed, and two kinds of greedy algorithms with greedy operator are presented. Then the simulated annealing algorithm based on greedy is proposed. To evaluate the performance of the mixed integer programming model, greedy algorithms and the simulated annealing algorithm based on greedy numerical experiments of the small-sized and large-sized problems are designed. Numerical experiments show that the modern mixed integer programming solver can solve some small size problems that the product of the number of machines and total jobs number is less than 75, simulated annealing algorithm based on greedy is effective and adapt to all kinds of size problems, and the interval insertion greedy algorithm can also obtain satisfactory results rapidly and suit for scenarios that need to be solved quickly.