改进布谷鸟算法求解双资源约束柔性车间调度问题

Improved cuckoo algorithm for flexible Job-Shop scheduling with dual resource constraints

针对双资源约束的柔性车间调度问题(DRCFJSP),以优化最大完工时间为目标,设计出一种具有改进解码方案的布谷鸟算法对其进行求解。由于DRCFJSP除了需要考虑机器的分配,还需要兼顾工人的加工情况,所以改进了传统解码方式以避免机器和工人在加工时间上的冲突,同时在解码时尽可能利用机器和工人的空闲时间。在布谷鸟算法核心框架下,将布谷鸟种群随机划分为三个子群,每个子群采用不同Lévy飞行方式独立进行寻优,并通过差分算子实现子群间信息交流,不仅增强了算法的全局搜索能力,也平衡了算法的局部搜索能力。最后通过基准测试算例进行实验仿真分析并与其他算法进行对比,验证了改进布谷鸟算法和改进解码方法的有效性和优越性。

Aiming at the flexible Job-Shop scheduling problem with dual resource constraints(DRCFJSP),this paper designed a cuckoo algorithm with an improved decoding scheme to solve it with the goal of optimizing the maximum completion time.Since DRCFJSP needs to consider the allocation of machines and the processing situation of workers,this paper improved the traditional decoding method to avoid the conflict of processing time between machines and workers,and made use of the idle time of machines and workers as much as possible during decoding.Under the core framework of the cuckoo algorithm,this paper divided the cuckoo population into three subpopulations randomly,and each subpopulation adopted different Lévy flight methods to search for the optimum independently,and realized the information exchange between subpopulations through the difference operator,which not only enhanced the global search ability of the algorithm,but also balanced the local search ability of the algorithm.Through the experimental simulation analysis of the benchmark test case,the results show the effectiveness and superiority of the improved cuckoo algorithm and the improved decoding method.