计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (14): 221-224.DOI: 10.3778/j.issn.1002-8331.2010.14.066

• 工程与应用 • 上一篇    下一篇

基于决策树的专用炉炉群生产策略挖掘

梁合兰1,李苏剑1,姚利苹2,麻卫平2   

  1. 1.北京科技大学 机械学院 物流研究所,北京 100083
    2.首钢迁安钢铁有限公司,河北 迁安 064404
  • 收稿日期:2008-12-24 修回日期:2009-03-16 出版日期:2010-05-11 发布日期:2010-05-11
  • 通讯作者: 梁合兰

Decision tree based production strategy mining for specific reheating furnaces mode

LIANG He-lan1,LI Su-jian1,YAO Li-ping2,MA Wei-ping2   

  1. 1.Department of Logistics Engineering,University of Science and Technology Beijing,Beijing 100083,China
    2.Hot Strip Mill of Shougang Qian’an Iron & Steel CO.,LTD.,Qian’an,Hebei 064404,China
  • Received:2008-12-24 Revised:2009-03-16 Online:2010-05-11 Published:2010-05-11
  • Contact: LIANG He-lan

摘要: 针对专用炉模式下的加热炉炉群生产要素间关系复杂,采用生产实践或仿真方法不易于找到适用的生产策略的不足,提出了一种应用决策树进行仿真结果分析,从而总结专用炉炉群生产策略的方法。该方法首先通过运行改进的专用炉模式下的炉群物流仿真模型获取训练样本集,然后采用决策树算法进行样本数据挖掘,最后通过决策规则分析总结出适用于专用炉模式的加热炉炉群生产组织策略。基于决策树的方法是专用炉炉群生产策略挖掘研究的新思路,研究结果显示该方法使仿真分析过程更简单和直观,并能挖掘出一些传统方法不容易被发现的生产规则。

关键词: 加热炉, 专用炉模式, 物流仿真, 决策树

Abstract: It is not easy to find effective production strategies for Specific Reheating Furnaces mode(SRF) by experiments or simulation methods due to the complexity in interrelations between production cells of SRF.In this paper,therefore,a decision tree based method to analyze the output of simulation model is proposed for production strategies mining of SRF.Firstly,an improved simulation model of SRF is proposed to obtain the training samples.Secondly,decision rules are mined out by means of decision tree algorithm using the acquired samples.Finally,production strategies of SRF are determined after decision rules analysis.The use of decision tree method is a novel approach in production strategies research of SRF,and results show that the method can make the decision analysis more simple and interpretable,and can mine out valuable rules including those which are not easy to be found.

Key words: reheating furnaces, specific reheating mode, logistics simulation, decision tree

中图分类号: