【作者】 郭钧；

【导师】 郭顺生；

【作者基本信息】 武汉理工大学 ， 工业工程， 2012， 博士

【摘要】 质量是制造企业占领市场最有力的战略武器,对企业的生存和发展至关重要。随着计算机技术、网络技术以及人工智能等技术的迅猛发展,使得整车制造企业生产模式和管理方法正发生着巨大的改变,传统的质量管理方法已经不能适应。因此,研究先进的质量管理方法和质量控制技术对于发展我国先进汽车制造模式具有极为重要的意义。论文以整车制造企业生产过程质量管理为背景,研究并分析了生产过程质量智能控制与诊断技术,主要研究内容及其具体工作如下：(1)结合汽车生产组织模式特征,分析了汽车生产过程质量控制特点和质量信息管理内容,提出了整车制造企业制造质量信息管理系统体系结构,构建了该体系下面向生产过程的质量信息集成模型,提出了整车制造企业生产过程质量控制研究的关键技术。(2)分析并指出了现有整车制造企业生产过程质量控制中的难点与不足。在此基础上,提出了基于改进极端学习机模型的生产过程质量智能预测方法,最后给出了该方法的实证分析。(3)提出了基于知识发现的生产过程质量数据挖掘模型。建立了面向生产过程的质量数据仓库,研究了关联规则提取的改进遗传算法和在质量数据仓库中的应用,并以整车制造企业应用实例进行了验证研究。(4)论文以过程质量持续改进为目标,研究了整车制造企业过程质量评价指标体系,利用数据包络分析方法建立了过程质量控制绩效的效率评价和诊断模型,并结合整车制造企业的实例对分析方法的有效性和实用性进行了验证。(5)论文最后从实例验证研究出发,针对案例整车制造企业生产过程质量管理的原型系统进行了需求与目标分析,给出该原型系统的总体框架、开发环境,开发并实现了原型系统,在案例企业中得的了具体应用,取到了良好的效果。更多还原

【Abstract】 Quality is the most powerful strategetic weapon of enterprise occupying market, which is very important to existence and development of enterprise. The production mode and management method of vehicle manufacturing enterprise is undergoing tremendous changes at the background of the rapid development of computer technology, network technology and artificial intelligence technology. Facing to the new changes of product quality management, conventional quality control method has not adapted to them. So, studying advanced method of quality management and quality control technology is especial and important significance to development vehicle manufacturing in china. In the background of vehicle manufacturing enterprise production process quality management, the problems of production process quality control and intelligent diagnosis technology are studied. The main tasks are as follows:(1) Combined with auto production pattern characteristics, process quality control model for vehicle manufacturing enterprise is put forward, system framework is constructed, function and architecture are determined on the basis.(2) The deficiencies of the existing process quality control methods in the production management of vehicle manufacturing enterprise are discussed. A process quality intelligent forecasting method combining Extreme Learning Machine (ELM) and Particle Swarm Optimization (PSO) is presented. Finally, a case study is given.(3) A production process quality data mining model based on knowledge extraction is presented. The quality data warehouse of production process is set up and a relevant decision-making knowledge extraction algorithm combining genetic algorithm (GA) and association rule (AR) mining is proposed.(4) To realize continual quality improvement, the process quality performance assessment model and indicators of vehicle manufacturing enterprise are set up and the evaluation and diagnosis models of quality performance efficiency are proposed by the use of Data Envelopment Analysis (DEA). The models and methods given by this paper are further illustrated by means of case study.(5) Finally, based on the instance validation study, especially for the process quality management in vehicle manufacturing enterprise, the requirement and objective for prototype system are analyzed, and then the overall framework and development environment are given. Furthermore, the prototype system is developed and used in a vehicle manufacturing company.更多还原