提出了一种基于人工神经网络（ＡＮＮ）的质量功能配置（ＱＦＤ）中顾客需求重要度评估方法．该方法的最大特点是直接从学习后的网络连接权重中提取出顾客需求重要度信息．讨论了网络的拓扑结构，以及如何从学习后的网络权重中提取出顾客需求重要度信息的方法．应用统计方法消除了网络学习初始权重对最终评估结果的影响．用一案例对提出的方法进行了验证，分析了网络隐含层节点数变化对顾客需求重要度评估结果的影响．实验表明，该方法具有很强的鲁棒性和适用性 This paper proposes an evaluation method of importance of customer’s needs in Quality Function Deployment (QFD) based on artificial neural network (ANN). The most important feature of this method is to extract the importance of customer demand directly from the weight of the network connection after learning. The topology structure of the network is discussed, and how to extract the importance information of customer needs from the learned network weights. The application of statistical methods to eliminate the initial weight of the network learning the final assessment results. A case is used to verify the proposed method and the influence of the change of nodes in the hidden layer of the network on the evaluation of the importance of customer needs is analyzed. Experiments show that this method has strong robustness and applicability