工业工程

工业工程 ›› 2012, Vol. 15 ›› Issue (5): 33-38.

• 实践与应用 • 上一篇    下一篇

考虑新任务插入的炼钢-连铸重调度模型与算法

侯东亮（1979-），男, 内蒙古自治区人，讲师，博士研究生，主要研究方向为生产计划与调度，智能优化算法.   

  1. (1.北京科技大学 东凌经济管理学院; 2.钢铁生产制造执行系统技术  教育部工程研究中心, 北京100083; 3. 广东海洋大学 工程学院，广东 湛江 524088)
  • 出版日期:2012-10-31 发布日期:2012-11-15
  • 基金资助:

    (教育部博士学科点专项科研基金资助项目(20100006110006)；中央高校基本科研业务费专项资金资助项目(FRF-AS-11-013B)

Rescheduling Model and Algorithm for Steelmaking and Continuous Casting with New Tasks Arrival

  1. (1.Dongling School of Economics and Management, University of Science and Technology Beijing, Beijing 100083, China;
    2.Engineering Research Center of MES Technology for Iron and Steel Production, Ministry of Education, University of Science and Technology Beijing, Beijing 100083, China;
    3.Engineering College, Guangdong Ocean University, Zhanjiang, 524088, China)
  • Online:2012-10-31 Published:2012-11-15

摘要： 针对新任务插入的炼钢-连铸重调度问题，将新任务插入映射为增加的时间约束，以开工时间差异化最小为目标建立了动态约束满足模型，提出了基于动态约束满足的局部修复算法。算法在变量选择中引入最小自由度优先策略用于选择变量，在值选择中提出最大空闲时间量可行分配策略用于给变量赋值，利用冲突识别规则识别赋值过程中产生的时序冲突和资源冲突并用冲突解消规则解消冲突。仿真实验模拟了3个具有代表性的插入时刻点，所得目标函数值均为零。结果表明所提算法能够最大限度地满足满足生产的实时性和稳定性的需求。

关键词: 调度修复, 动态约束满足, 重调度, 炼钢连铸

Abstract: A rescheduling problem of steelmaking-continuous casting with new tasks arrival was studied in this paper. By mapping new tasks that would be insert an initial schedule to the added time constraint, a dynamic constraint satisfaction model was established to minimize the difference of the start time of the old jobs between the initial schedule and the adjust one. And according to this model, a local repair rescheduling algorithm based on dynamic constraint satisfaction techniques was put forward. In this method, the Little Freedom First(LFF) strategy was used to select variables and the Biggest Idle time Feasible Distribution(BIFD) strategy to assign one value to a variable. The conflict identification and elimination principles were used to identify and eliminate the time conflicts and resource conflicts in the assignment. In simulation experiment, three different insertion time points were generated randomly. Target value were all zero. The results demonstrate that the algorithm can satisfy the needs of realtime and stability as much as possible.

Key words: schedule repair, dynamic constraint satisfaction, rescheduling, steelmaking and continuous casting