中国机械工程 ›› 2014, Vol. 25 ›› Issue (16): 2191-2195,2203.

• 智能制造 • 上一篇    下一篇

基于混合顺序启发式算法的一维下料问题

程浩;刘心报;方昶   

  1. 合肥工业大学,合肥,230009
  • 出版日期:2014-08-26 发布日期:2014-09-11
  • 基金资助:
    国家自然科学基金资助重点项目(71231004)

A Hybrid Sequential Heuristic Procedure to One-dimensional CSP

Cheng Hao;Liu Xinbao;Fang Chang   

  1. Hefei University of Technology,Heifei,230009
  • Online:2014-08-26 Published:2014-09-11
  • Supported by:
    National Natural Science Foundation of China(No. 71231004)

摘要:

针对一维下料问题,提出一种改进的混合顺序启发式算法。该算法修正了传统顺序启发式算法的待切子集选择策略,采用广度搜索算法在有限时间内解决终止条件问题,根据用户最大时间容忍度动态地改变参数。最后采用C#进行编程,实现了该算法,通过对比证明算法的有效性。该算法在马鞍山钢铁公司车轮分公司正式使用了两年多,使切割率提高4%。

关键词: 下料问题, 动态规划, 顺序启发式算法, 背包问题

Abstract:

This paper proposed an improved hybrid sequential heuristic procedure for one-dimension CSP.The algorithm modified the subset selection strategy of original sequential heuristic procedure and avoided the ending condition in the limited time with width search algorithm,and could change the parameters according to the users' maximum time patience dynamically.Finally C# was used to realize the algorithm.the results show the algorithm is effective.This algorithm were used in Ma'anshan Steel Corp,wheel company official for more than two years,and increases  the cutting rate 4%.

Key words: cutting stock problem(CSP), dynamic programming, sequential heuristic procedure, knapsack problem

中图分类号: