项目名称: 失效控制下石化企业批量生产计划与设备维修协同决策模型
项目编号: No.71471116
项目类型: 面上项目
立项/批准年度: 2015
项目学科: 管理科学
项目作者: 吕文元
作者单位: 上海理工大学
项目金额: 60万元
中文摘要: 石化企业批量生产计划与设备维修协同决策的研究,旨在确保石化企业安全生产,减少高额的维修费用,以及满足市场对石化柔性生产需求。首先,建立基于部分可观测信息的隐半马可夫模型,结合贝叶斯图法对关键设备失效控制,预测出设备的剩余寿命。其次,在保证设备安全性的前提下,根据石化批量生产的特点,建立批量生产计划与维修协同决策模型,制定最优化的生产批量、批次和预防维修间隔期。并建立基于嵌入式的吸收态马尔可夫链的2M1B(Two-machine one-buffer)模型,解决石化复杂物流、大量中间在制品库存下的生产计划与控制、在制品库存和上下流设备维修策略综合决策的问题,确保生产平准化。最后开发智能的生产与维修协同决策支持系统,提高管理人员的生产与维修决策水平。这些研究将设备维修管理带入一个新的发展阶段,丰富了生产管理理论,对于提高石化企业的生产管理水平和经济效益都具有重要的理论意义和实用价值。
中文关键词: 生产计划;设备维修;寿命预测;隐半马尔可夫
英文摘要: The research is about modelling of simultaneous determination of batch production plan and maintenance schedule for petrochemical enterprise, its objective is to guarantee the safety of the production line,reduce the highly maintenance cost, meet the need of flexible production. Firstly , the Hidden semi-Markov model,HSMM, based on partial observable information is established to predict the useful residual life of the petrochemical equipment and further to combine the Bayesian method to guarantee the safety of petrochemical production system. Secondly, synchronous decision model is presented to simultaneously determine the production planning and maintenance schedule, which is suitable to the features of petrochemical production. Using the embedded absorbing Markov chain, the model of Two-machine one-buffer, 2M1B, is proposed to determine the optimal buffer, the production plan and the maintenance policy. Finally, the intelligent decision support system is presented to help the decision maker to make the production planning and maintenance schedule. The achievements of this study have greatly theoretical values not only on the increase of the level of production management and the economic efficiency of petrochemical enterprises, but also on the enrichment of the maintenance theories.
英文关键词: production plan;equipment maintenance;residual life prediction;hidden semi-Markov model