Job-shop scheduling using genetic algorithm
Job-shop scheduling using genetic algorithm
复制标题
DOI：
10.1109/icsmc.1996.565434
复制
发表时间：
1996-10
期刊：
影响因子：
--
通讯作者：
Wu Ying;Li Bin
中科院分区：
文献类型：
--
作者：
Wu Ying;Li Bin
车间作业调度是CIMS环境下计划和生产控制的重要环节，是一个典型的NP完全问题。车间调度的研究主要集中在基于知识的方法和启发式搜索，这是有用的，除了知识的获取困难。遗传算法是一种利用自然界进化思想的优化方法。遗传算法虽然简单，但效率很高。本文提出了三种新的遗传算法模型，即十进制空闲时间编码遗传算法（DITCGA）、二进制空闲时间编码遗传算法（BITCGA）和自适应空闲时间编码遗传算法（AITCGA），用于设计车间作业调度算法。利用空闲处理时间对问题进行编码，有效地减少了求解空间。在我们的方法中，一个自适应学习机制，以指导搜索或进化过程。仿真结果表明了这些方法的有效性。
Job-shop scheduling, a typical NP-complete problem, is an important step in planning and manufacturing control of CIMS environments. Researches on job-shop scheduling focus on knowledge-based approaches and heuristic searching which are useful apart from the difficulty of obtaining knowledge. Genetic algorithms are optimization methods which use the ideas of the evolution of nature. Simple as genetic algorithms are, they are efficient. Three novel genetic algorithms models, such as decimal idle time coding genetic algorithms (DITCGA), binary idle time coding genetic algorithms (BITCGA), and adaptive idle time coding genetic algorithms (AITCGA), are presented to design a job-shop scheduling algorithm in this paper. Using the idle processing time to code this problem, we efficiently reduce the solution space. In our approaches, an adaptive learning mechanism is applied to guide the searching or evolution process. The simulation results show the efficiency of these approaches.