协同研制背景下复杂装备可靠性增长GERT网络模型研究

The complex equipment multiple cooperating system development process generally adopts the main manufacturer-supplier management mode. A lot of research and literature study found that existing reliability growth model cannot provide a better solution for the equipment reliability growth evaluation and prediction. In this project, from the view of the characteristics of modern complex equipment, we set up a complex equipment reliability growth GERT network framework under the background of collaborative development to research the reliability growth mechanism.Secondly, on the basis of GERT network, equipment reliability growth test data “equivalent transfer” model, Bayes reliability growth evaluation inference model, equipment failure rate, improve effectiveness coefficient and growth potential of the Bayes network estimate model, common mode failure under the condition of reliability growth evaluation correction model are established. Thirdly, considering each supplier for the responsibility of the equipment reliability growth, the difference between the position and contribution and resource allocation problems under the background of collaborative development, we set up equipment supplier identification and evaluation of reliability growth model based on GERT network, resource optimal allocation model.Finally, with the background of China C919 development process, theoretical models in the project are tested. New solution for the collaborative development under the background of complex equipment reliability growth model and technology development is provided.

复杂装备的多系统协同研制过程一般采用主制造商－供应商组织管理模式，大量的调研和文献研究发现，现有的可靠性增长模型无法较好地解决该模式下装备可靠性增长评价与预测等相关问题。本项目首先从现代复杂装备研制特征出发，搭建多系统协同研制背景下复杂装备可靠性增长GERT网络框架，解析其可靠性增长机制；其次，建立基于GERT网络的装备可靠性增长试验数据“折合传递”模型，可靠性增长评价Bayes推理模型，装备失效率、改进有效性系数和增长潜力的Bayes网络估算模型，共模故障条件下的可靠性增长评价修正模型；再者，考虑协同研制背景下各供应商对于装备可靠性增长的责任、地位与贡献的区别与资源配置问题，建立基于GERT网络的装备可靠性增长供应商识别与评价模型，资源最优配置模型；最后，以中国C919飞机研制为背景，对该项目的理论模型进行检验与应用；为协同研制背景下复杂装备可靠性增长模型建立与技术开发提供解决方案。

复杂装备的研制大多具有“主制造商-供应商”协同研制的特点，装备的可靠性增长过程作为研制的关键环节融于其生产全过程，具有分布式网络化协同特征，且获得的数据多源异构。本课题吸纳相关领域的最新研究成果，综合运用概率论、随机过程、GERT网络技术、Bayes推理与分析方法、灰色系统理论等先进工具，在灰信息背景下的复杂装备可靠性增长AMSAA模型、小子样背景的复杂装备可靠性寿命试验RLT-GERT网络模型、协同研制背景下复杂装备可靠性增长FTA-GERT网络模型、基于多智能体系统的共因失效系统故障预测GERT网络模型、分层管控架构下复杂装备研制项目进度计划模型、考虑老化效应的系统竞争失效分析与薄弱环节识别、面向质量竞争力的复杂装备研制供应链质量成本管控模型、相关竞争失效系统可靠性建模及其预防性维修决策优化等方面取得重要进展，在新的视角下为协同研制背景下复杂装备可靠性增长评价与预测问题提供系统、科学的解决方案。同时积极将理论模型运用到以我国C919大型民用飞机研制为代表的可靠性增长案例分析中去，服务国家重大需求。