交通运输系统工程与信息 ›› 2012, Vol. 12 ›› Issue (3): 129-134.

• 系统工程理论与方法 • 上一篇    下一篇

基于模糊粗糙集与支持向量机的区域物流量预测

何满辉*,逯林,刘拴宏   

  1. 辽宁工程技术大学 教务处,辽宁 葫芦岛 125105
  • 收稿日期:2012-01-11 修回日期:2012-03-12 出版日期:2012-06-24 发布日期:2012-07-03
  • 作者简介:何满辉(1973-),男,山西绛县人,副教授,博士.
  • 基金资助:

    国家自然科学基金项目(70971059);辽宁省教育厅人文社会科学研究项目(2009A342).

Forecasting Regional Logistics Amount Based on FuzzyRough Set and SVM

HE Man-Hui, LU Lin, LIU Shuan-Hong   

  1. Dean’s Office, Liaoning Technical University, Huludao 125105, Liaoning, China
  • Received:2012-01-11 Revised:2012-03-12 Online:2012-06-24 Published:2012-07-03

摘要:

对区域物流量进行研究与预测有助于把握区域物流的需求,实现区域物流供需相对平衡,提高区域物流规划质量和运行效率具有重要的理论和实际意义.本文将模糊粗糙集理论引入区域物流量的预测中,建立基于模糊粗糙集与支持向量机的区域物流量预测模型,用模糊粗糙集作为前端预处理器对数据进行约简,剔除冗余信息,以实现两种算法的优势互补.针对支持向量机在处理数据时无法将数据简化的问题,提出了基于模糊粗糙集与支持向量机的区域物流量预测方法,在支持向量机对样本数据进行处理之前,利用模糊粗糙集数据挖掘的能力对原始数据样本集进行预处理.结果表明,这种预测方法具有很好的精确性和有效性.

关键词: 物流工程, 支持向量机, 模糊粗糙集, 物流预测

Abstract:

The regional logistics research and forecasting is helpful for grasping the regional logistics demand, enforce relative balance between supply and demand of regional logistics and improve the quality of regional logistics planning and operational efficiency. To realize the complementary advantages of two algorithms, this paper introduces the fuzzyrough set theory into regional logistics forecast. It develops a model for regional logistics forecasting based on the fuzzy rough set and SVM. It then applies the fuzzy rough set as a frontend processor for data reduction and eliminates the redundant information. Because the SVM for processing data in the data reduction cannot be easily simplified, this paper proposes a forecasting method of regional logistics based on the fuzzy rough set and SVM. Before processing the sample data by SVM, the data is used to enlarge the capabilities of fuzzy rough set on original data samples for preprocessing. The results show that this prediction method has good accuracy and effectiveness.

Key words: logistics engineering, support vector machine (SVM), fuzzrough set, logistics forecasting

中图分类号: