Abstract: In order to improve the selection quality of knowledge service in cloud manufacturing environment, a cloud manufacturing knowledge service optimization method based on user’s psychological behavior is proposed. Based on the analysis of cloud manufacturing knowledge service characteristics, the optimization evaluation index system of cloud manufacturing knowledge service is set up. The rough set theory is used to assign initial weights to each evaluation index, and the initial weights are adjusted according to the multiple-attribute preferences of user’s to ensure the rationality of weight distribution. On this basis, combining with the users psychological behavior analysis in the decision-making process and regret theory, the perceived value matrix of cloud manufacturing knowledge service is constructed, and the comprehensive perceived value of each knowledge service is calculated by integrating the above index weights to obtain the optimal cloud manufacturing knowledge service. Finally, an application example is used to verify the effectiveness and practicality of this method.

Key words: cloud manufacturing, knowledge service, regret aversion, psychological behavior, multiple-attribute preference

摘要： 为提高云制造环境下知识服务的选择质量，提出一种基于用户心理行为的云制造知识服务优选决策方法。基于云制造知识服务的特性分析，建立云制造知识服务优选评价指标体系；运用粗糙集理论，对各评价指标进行初始权重分配，并根据用户多属性偏好对初始权重进行调整，以确保权重分配的合理性；在此基础上，结合用户在决策过程中的心理行为分析和后悔理论，构建云制造知识服务的感知价值矩阵，并计算各知识服务的综合感知价值，以获得最优云制造知识服务。通过一个应用实例，验证此方法的有效性和实用性。

关键词: 云制造, 知识服务, 后悔规避, 心理行为, 多属性偏好