上海交通大学学报

上海交通大学学报（自然版）

• 管理科学 • 上一篇    下一篇

基于支持向量机的武器装备研制项目风险评价方法

李勘1,2   

  1. (1. 西北工业大学 机电学院， 西安 710072； 2. 海军大连舰艇学院 科研部， 大连 116018)
  • 收稿日期:2007-11-29 修回日期:1900-01-01 出版日期:2008-11-28 发布日期:2008-11-28
  • 通讯作者: 李勘

Risk Evaluation Method of Weapon Equipment Development Project Based on Support Vector Machine

LI Kan1,2   

  1. (1. School of Mechanical and Electronical Engineering, Northwestern Polytechnic University,
    Xi’an 710072, China; 2. Department of Sciences Research, Dalian Naval Academy, Dalian 116018, China)
  • Received:2007-11-29 Revised:1900-01-01 Online:2008-11-28 Published:2008-11-28
  • Contact: LI Kan

摘要： 针对武器装备研制项目风险评价方法在实际应用中存在的问题，对支持向量机在武器装备研制项目风险评价中应用的可行性进行了分析，提出了基于支持向量机的武器装备风险评价模型.应用收集相关项目的数据资料进行了实证研究，并与神经网络和模糊综合评价法进行了比较分析.结果表明，基于支持向量机的评价模型具有良好的自学习性和特征提取能力，可为武器装备研制项目风险评价提供有益的参考.

关键词: 支持向量机, 武器装备研制项目, 风险评价

Abstract: Aiming at the problems in application of the evaluation method of weapon equipment development project, a risk evaluation model of weapon equipment development project based on support vector machine was proposed based on the explanation of the theory of support vector machine(SVM) and analysis of the feasibility of SVM in the risk evaluation of weapon equipment development project. The predicated results were verified by the project data, which shows a good agreement. Comparing SVM model with neural network and fuzzy comprehensive evaluation method shows that the SVM model has better ability of selflearning and feature extraction, and can provide a helpful suggestion for the risk evaluation of weapon equipment development project.

中图分类号: