华南理工大学学报(自然科学版) ›› 2013, Vol. 41 ›› Issue (1): 89-94.doi: 10.3969/j.issn.1000-565X.2013.01.014

• 计算机科学与技术 • 上一篇    下一篇

基于人工免疫机理和LS-SVM 的顾客需求重要度预测

黄爱华1 蒲洪彬2 李伟光2 侯跃恩2   

  1. 1. 华南理工大学 工商管理学院, 广东 广州 510640; 2. 华南理工大学 机械与汽车工程学院, 广东 广州 510640
  • 收稿日期:2012-07-05 修回日期:2012-09-21 出版日期:2013-01-25 发布日期:2012-12-03
  • 通信作者: 黄爱华(1962-),女,副教授,主要从事管理科学与工程、企业管理研究. E-mail:ahhuang@scut.edu.cn
  • 作者简介:黄爱华(1962-),女,副教授,主要从事管理科学与工程、企业管理研究.
  • 基金资助:

    广东省重大科技专项(2009A080202004);广东省粤港关键领域重点突破项目(2011BZ100012)

Importance Forecast of Customer Requirements Based on ArtificialImmunity and Least Square Support Vector Machine

Huang Ai-hua1 Pu Hong-bin2 Li Wei-guang2 Hou Yue-en2   

  1. 1. School of Business Administration, South China University of Technology, Guangzhou 510640, Guangdong, China;2. School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou 510640, Guangdong, China
  • Received:2012-07-05 Revised:2012-09-21 Online:2013-01-25 Published:2012-12-03
  • Contact: 黄爱华(1962-),女,副教授,主要从事管理科学与工程、企业管理研究. E-mail:ahhuang@scut.edu.cn
  • About author:黄爱华(1962-),女,副教授,主要从事管理科学与工程、企业管理研究.
  • Supported by:

    广东省重大科技专项(2009A080202004);广东省粤港关键领域重点突破项目(2011BZ100012)

摘要: 为了提高质量功能配置过程的动态决策水平,实现顾客需求重要度的有效预测,采用最小二乘支持向量机(LS-SVM) 建立顾客需求重要度预测模型,并结合人工免疫机理,以LS-SVM 模型参数为抗体、顾客需求重要度预测误差为抗原,采用克隆算法完成顾客需求重要度LS-SVM 预测模型参数的免疫优化选择. 文中还以数控瓦楞纸板印刷模切机顾客需求重要度预测为例验证了所提方法的可行性,并与基于灰色模型和神经网络模型的顾客需求重要度预测方法进行了比较,结果表明,所提出的基于人工免疫机理与LS-SVM 的顾客需求重要度预测方法可行,且预测效果较好.

关键词: 质量功能配置, 顾客需求, 重要度预测, 人工免疫, 最小二乘支持向量机

Abstract:

In order to improve the dynamic decision level during the quality function deployment and effectivelypredict the importance of customer requirements, a forecast model based on the least square support vector machine(LS-SVM) is constructed. Then, by using the model parameters as the antibody and the importance forecast devia-tion as the antigen, the parameters of LS-SVM model are optimized by means of the clonal selection algorithm. Fi-nally, a numerical printing and die-cutting machinery of corrugated board is used as an example to verify the feasi-bility of the proposed method, and the results are compared with those obtained by the methods based on grey modeland neural network. It is found that the proposed method based on artificial immunity and LS-SVM is feasible andeffective in forecasting the importance of customer requirements.

Key words: quality function deployment, customer requirement, importance forecast, artificial immunity, least-square support vector machine

中图分类号: