摘要
有效地混合了遗传算法和基于约束满足的自适应神经网络算法,对于一类加工时间可变的调度问题进行了研究.遗传算法被用来进行迭代寻优.当前代经交叉和变异后生成的染色体对应非可行解,由自适应神经网络运算后得到可行解,对应的染色体作为新一代染色体.本算例的目标函数是基于任务的提前/拖期惩罚、附加惩罚以及加工时间的偏离量惩罚,目标是确定最优加工时间和最优加工顺序极小化目标函数,并与一般的遗传算法相比较,实验结果说明了遗传/自适应神经网络算法混合算法的有效性.
A new algorithm which effectively combines genetic algorithm （GA） and constraint satisfaction adaptive neural network （CSANN） was proposed to solve job-shop scheduling problems where the jobs have variable processing times. In the hybrid method, GA is used to iterate for searching optimal solutions, CSANN is used to solve feasible solutions during the iteration of GA. The total objective function is based on earliness/tardiness penalties, additional penalties and the deviation of processing times penalties. The objective is to find the optimal common due - date, the opt imal sequence and optimal processing times to minimize the objective function. Computer simulations have shown the good performance of the proposed hybrid method.
出处
《赣南师范学院学报》
2007年第6期66-68,共3页
Journal of Gannan Teachers' College(Social Science(2))
关键词
遗传算法
自适应神经网络
可变加工时间
调度
genetic algorithm
adaptive neural network
variable processing times
scheduling
