富信息环境下复杂可修系统动态维修决策研究

项目名称： 富信息环境下复杂可修系统动态维修决策研究

项目编号： No.71501103

项目类型： 青年科学基金项目

立项/批准年度： 2016

项目学科： 管理科学

项目作者： 周瑜

作者单位： 内蒙古大学

项目金额： 17.5万元

中文摘要： 因生产系统突发故障的后果十分严重，故障预测与健康管理、基于状况的维修、决策支持系统等维修理念和信息化技术得到了丰富地研究及应用。这就为系统的维修管理创造了一个数据和信息丰富的环境，即富信息环境。然而，面对数据源的多样性和复杂性，管理者很难综合利用这些数据辅助维修管理决策，大多局限于对实时监测、失效时间间隔和费用等常规数据的分析和利用。特别是在动态维修管理决策中，缺乏对多样性数据的融合与分析。因此，本项目将综合运用数理统计、数据挖掘和多目标决策等技术，尝试融合多源、多时相和多维数据，分析这些数据与系统失效之间的潜在规律，研究离线和在线信息数据间的关联规则，为动态维修决策提炼信息。在此基础上，开发动态维修管理决策新模型新方法，构建富信息环境下复杂系统维修最优化的理论框架。本研究将为应用多样性数据辅助开展复杂系统动态维修决策提供理论依据。

中文关键词： 可修系统；维修决策优化；成组维修；机会维修；富信息

英文摘要： Advanced maintenance philosophy and information technology, such as prognostics and health management, condition-based maintenance and decision support systems, have been studied and applied since the sudden failure could bring out some very serious consequences. So this creates a rich data and information environment for system maintenance management, as called rich-information environment. However, faced with the diversity and complexity of data sources, the managers are very difficult to utilize these data to assist maintenance management decisions. And only partly of the regular data are used, such as the real-time monitoring data, time between failures and cost. Especially in the dynamic maintenance management decisions, the lack of the integration and analysis of the diversity management data is more serious. Therefore, this study will try to merge the multi-source, multi-temporal and multi-dimensional data, analyze the potential discipline between these data and system failure, research the association rules between online and outline data and extract information for dynamic maintenance decision through applying the mathematical statistics, data mining and multi-objective decision-making techniques. On this basis, this study will develop new models and methods of dynamic maintenance management decisions and build the complex systems maintenance optimization theoretical framework under rich-information environment. This study also will provide a theoretical foundation for assisting the complex system dynamic maintenance decision with using the diversity data.

英文关键词： Repairable system;Maintenance decision optimization;Group maintenance;Opportunity maintenance;Rich-information