以利润和熵为性能指标的车间生产调度

Job Shop Scheduling With Profit and Entropy as Performance Measures

为解决以往研究中车间调度指标考虑因素不全面的问题,将时间、质量、成本、能耗和环境5个因素统一为利润指标、提出了以利润作为调度指标、熵作为调度方案有效性的评估指标、基于遗传算法的柔性制造车间调度方法.该方法以利润指标作为遗传算法的染色体适应值,经多次寻优,求解出一组次优调度方案,对次优调度方案集进行利润与熵为指标的基于熵权的多目标决策,决定最终的最优调度方案.应用实例和程序进行验证,结果表明,与传统调度方法相比,所提出的方法在指标的全面性及实用性上有一定优势.

To solve the unilateral problems of job-shop scheduling performance measures used in previous studies,by using an unified measure-profit integrated time,quality,cost,energy and environment along with entropy as evaluating the scheduling effectiveness,an approach for job-shop scheduling algorithm is proposed based on genetic algorithm. Wherein,the profit is regarded as the adaptive value of the chromosome in genetic algorithm,a set of suboptimal scheduling schemes gained via searches,and the best scheduling scheme is gained after decision-making for the suboptimal scheduling schemes with profit and entropy as objectives summed up by entropic weights. Examples and programs in program language C# are used to illustrate validation. It shows that the proposed approach have some advantage than that of traditional scheduling algorithms in all sidedness and practicability of performance measures.