摘要
为了解决客户协同创新中协同工作效率难于评价的问题,提出了一种用遗传算法优化的神经网络对客户协同产品创新进行评价的评价模型:在评价指标方面,设计了一套包括效益、效率和过程的18个指标的评价体系;在评价算法方面,将遗传算法与BP神经网络结合起来,设计了遗传算法改进的BP神经网络算法。该模型充分利用遗传算法的全局搜索能力强与神经网络的局部搜索能力强的特点,克服了遗传算法局部收敛与神经网络收敛速度较慢的问题,是一种非常适用于评价协同工作的模型。最后通过实例训练,证明了该模型的有效性与可行性。
In order to solve problem of efficiency evaluation of collaborative work in customer collaborative innovation,The evaluation model which uses neural network based on genetic algorithm to evaluate customer collaborative product innovation was presented.In the evaluation indicators,we designed a set of evaluation system which included 18 indicators.In the evaluation algorithms,we combined BP neural network and genetic algorithm,and designed a set of BP neural network algorithm improved by genetic algorithm.The algorithm made fully use of genetic algorithm global searching and BP network local searching,overcomed the local convergence of genetic algorithm and lower efficiency of neural network convergence.Finally,a numerical example was use to illustrate the feasibility and availability of the evaluation model.
出处
《机械》
2010年第8期5-9,共5页
Machinery
基金
教育部高校博士点科研基金资助项目(20090191110004)
关键词
遗传算法
BP神经网络
客户协同创新
genetic algorithm
BP neural network
customer collaborative innovation
