摘要
论述了工时定额的概念及影响因素、常用的工时定额制定方法、线性神经网络模型的不足。提出了一种模糊极大极小人工神经网络模型的工时定额测算方法,该方法不仅具有一般神经网络的优点,可以输入n维模糊量,而且能够直接处理模糊变量。同时,网络的训练采用无师学习方法,使得模型能够应付新出现的影响因素,从而可以解决新技术、新工艺的出现对工时定额的动态影响。
Conception and influence factors of time quota estimation, traditional algorithm and shortcoming of linearity neural network were discussed. A kind of algorithm based on the fuzzy min-max artificial neural network model is put forward. It not only has the advantages of normal neural network, but also can be used to deal with fuzzy variables. Meanwhile, an unsupervised learning method is used to train the network, so new influence factors can be coped with by this model, and the dynamic effects of the appearance of new technique and craftwork to time quota can be solved.
出处
《公路交通科技》
CAS
CSCD
北大核心
2007年第2期155-158,共4页
Journal of Highway and Transportation Research and Development
基金
河南省交通科技资助项目(2005P451)
关键词
公路工程
工时定额
神经网络
无师训练
highway engineering
time quota
neural network
unsupervised training