工业工程

工业工程 ›› 2018, Vol. 21 ›› Issue (6): 7-15,22.doi: 10.3969/j.issn.1007-7375.2018.06.002

• 实践与应用 • 上一篇    下一篇

基于本体案例匹配的扰动作业车间智能调度辅助决策

吴正佳1,2, 涂晶鑫2, 华露2, 白炜铖2, 刘秀凤2   

  1. 三峡大学 1.水电机械设备设计与维护湖北省重点实验室;
    2. 机械与动力学院, 湖北 宜昌 443002
  • 收稿日期:2018-06-05 出版日期:2018-12-30 发布日期:2018-12-29
  • 作者简介:吴正佳(1964-)，男，湖北省人，教授，博士，主要研究方向为生产调度及智能算法应用
  • 基金资助:
    国家自然科学基金资助项目（51641505）

A Research on Intelligent Scheduling Decision for Disturbance Job Shop Based on Ontology-CBR

WU Zhengjia1,2, TU Jingxin2, HUA Lu2, BAI Weicheng2, LIU Xiufeng2   

  1. 1. Hubei Key Laboratory of Hydroelectric Machinery Design & Maintenance;
    2. College of Machanical and Power Engineering, China Three Gorges University, Yichang 443002, China
  • Received:2018-06-05 Online:2018-12-30 Published:2018-12-29

PDF

5807

摘要： 为了快速响应扰动事件，恢复生产稳定，创新性地提出了多类型扰动事件本体与车间生产调度案例本体智能匹配技术。通过引入本体技术（ontology technology），将以往生产动态调度的成功案例进行规范化和标准化处理，构建动态调度本体化的成功案例库和案例库自我完善的学习机制。结合案例推理（case-based reasoning，CBR）技术及相似度理论，将新扰动的生产车间调度问题与案例库的成功案例进行匹配，依据相似度所在阈值区间，实现扰动车间智能调度辅助决策，从而达到提高决策过程时效性，缩短决策时间的目的。通过仿真结果表明：与传统CBR及混合驱动调度策略相比，在本体-CBR的方法下比传统CBR的案例检索精度大约提高了6%，比混合驱动调度策略相比，在决策时间上要快25 min，故有效地提高了案例检索的精度与响应时间。

关键词: 本体, 案例推理(CBR), 生产扰动, 动态调度, 智能决策

Abstract: In order to respond quickly to disturbance events and restore production stability, the intelligent matching technology of multi-type perturbation event ontology and workshop production scheduling case ontology were proposed. Firstly, the successful cases of dynamic scheduling in the past were standardized through the introduction of ontology technology, and the successful ontological case libraries and the self-improvement learning mechanism of the case libraries constructed. Then, the problem of the new perturbed production shop scheduling with the success stories of the case database were matched by Case-Based Reasoning (CBR) technology and similarity theory, and perturbation workshop intelligent auxiliary decision actualized based on the similarity degree of the threshold interval, which achieved the purpose of improving the decision-making process timeliness and shortening the decision-making time. Finally, the simulation results show that, compared with the traditional CBR and hybrid-driven scheduling strategy, the case retrieval accuracy of the ontology-CBR method is about 6% higher than that of the traditional CBR, and the decision time is 25 mins faster than that of the hybrid-driven scheduling strategy, so the case retrieval accuracy and response time are effectively improved.

Key words: ontology, case-based reasoning (CBR), production disturbance, dynamic scheduling, intelligent decision

中图分类号: