基于广义产品知识的决策支持系统关键技术研究

发布时间：2018-04-16 12:25

本文选题：工业设计 + 决策支持系统　；参考：《西北工业大学》2014年博士论文

【摘要】：随着市场竞争环境的日益激烈，面对要求越来越高的产品质量、制造成本和售后服务等指标，迫使设计人员和制造企业要在更短的产品开发和生产周期内，以更低的成本制造出更为精良的产品。70%～80%的生产成本是在设计阶段就已决定的，因此产品的设计决策正确性对制造成本具有巨大的影响。而新的网络技术和信息技术的不断涌现，在全球范围内造就了一个利用知识来创造新价值的新经济时代，企业之间的竞争也逐渐转变为以知识为中心的竞争。一方面，在产品开发及制造的过程中产生了大量以指数形式增长的产品知识，相关产品知识呈现知识异构、分布式和动态性等特点，造成产品知识的获取与表示的困难，这一问题阻碍了知识的共享和重用，使得以知识为基础的决策支持系统效率大打折扣。另一方面，随着决策规模的扩大，设计决策活动的频率加快，面对网络环境下存在的大量丰富决策资源，如何在丰富的网络决策资源中选择最有能力的决策模型，，为产品开发中的设计决策和推理服务，已成为一项迫切而重要的研究课题。提高产品设计决策的质量离不开知识，而知识的有效利用又与基于知识的决策方法密切相关。本文研究的核心是以本体论和语义技术为基础，以知识工程的知识处理方法为主线，探索获取产品设计及制造阶段知识的有效方法，提出基于本体的知识建模方法和原则，明确决策过程中知识的评价方法和作用原理等，阐明决策支持的过程、方法和技术路线，详细论述了决策支持系统的运作模式及决策任务求解方法，据此建立并完善基于广义产品知识的决策支持系统的理论体系，并在此基础上，利用软件开发方法，设计出较为可行的，以提高决策者决策水平、效率和能力为目标的基于本体的广义产品知识的GPKDSS原型系统。论文的主要内容和创新点如下： 1、运用系统集成的思想，提出了基于广义产品知识的决策支持系统理论，设计了决策支持系统的体系结构，建立了初步的层次模型、逻辑模型框架以及决策支持系统详细结构。指出了对于广义产品知识决策所面临问题的解决思路，并对其中的关键技术进行了研究和实现。 2、在对决策支持系统需求分析的基础上，提出了“知识是基础，决策是目的，评价是辅助，语义贯穿于整个基于知识的决策支持系统中”这一基本思路，构成了决策支持系统的技术体系。 3、在分析产品的全生命周期的理论基础上，提出了广义产品知识的概念，研究了基于Internet的广义产品文本知识获取方法和基于知识地图的广义产品服务知识获取方法，重点解决了系统内产品知识的获取和有效信息的获取效率等问题；提出了产品知识评价指标体系的确定原则，在此基础上构建了知识综合评价指标体系，确定了指标的量化计算方法，完善了应用优度综合评价法对产品知识进行评价的理论框架，并以实例验证了该方法的可行性。 4、将本体工程与知识建模领域的研究相结合，建立了基于本体的知识建模原则，提出一种面向广义产品设计知识的本体构建方法，为所有产品知识资源提供统一的描述标准，并以某车型为例，构建了知识本体模型框架，描述了主要的类和类属关系以及专用罐式车知识本体OWL描述片段。 5、利用面向服务架构和语义Web技术，提出了决策任务提交及分解算法，构建了决策任务相互调用协作的运作流程；提出了基于语义的决策服务模糊聚类算法，并以实例验证了该方法的可行性。 6、论述了系统开发策略与功能模块划分以及系统开发工具的选择原则，开发了基于广义产品知识的决策支持系统的原型系统GPKDSS，并以某特种车辆生产企业为实例，对论文提出的理论和方法进行了验证。
[Abstract]:With the increasingly fierce market competition environment , facing the demand of higher product quality , manufacturing cost and after - sales service , the designers and manufacturing enterprises are forced to produce more sophisticated products at lower cost during the shorter product development and production cycle . 70 % -80 % of the production cost is determined in the design phase . Therefore , the correctness of the design decision - making of the product has a great impact on the manufacturing cost .

On the one hand , in the process of product development and manufacturing , a lot of product knowledge has been generated in exponential form , the knowledge of related products presents the characteristics of heterogeneous , distributed and dynamic , which hinders the acquisition and representation of product knowledge . This problem impedes the sharing and reuse of knowledge so that the efficiency of the knowledge - based decision support system is greatly reduced .

On the other hand , with the enlargement of the scale of decision - making , the frequency of designing decision - making activities is speeding up , and it is an urgent and important research topic how to select the most powerful decision - making model in the rich network decision - making resources in the face of the abundant decision - making resources in the network environment .

This paper discusses the process , method and technical route of decision support based on ontology and semantic technology .

The main content and innovation points of the thesis are as follows :

1 . Based on the idea of system integration , a decision support system theory based on generalized product knowledge is put forward . The architecture of the decision support system is designed , and a preliminary hierarchy model , a logical model framework and a detailed structure of the decision support system are established .

2 . Based on the analysis of the demand of DSS , the paper puts forward " knowledge is the foundation , the decision is the aim , the evaluation is auxiliary , the meaning runs through the whole knowledge - based decision support system " , which constitutes the technical system of the decision support system .

3 . Based on the analysis of the whole life cycle theory of the product , the concept of generalized product knowledge is put forward , and the knowledge acquisition method of generalized product text based on Internet and the knowledge acquisition method of generalized product service based on knowledge map are studied , and the problems such as the acquisition of product knowledge in the system and the acquisition efficiency of effective information are solved .
This paper puts forward the principle of determining the index system of product knowledge evaluation , on the basis of which , a comprehensive evaluation index system of knowledge is constructed , the quantitative calculation method of the index is determined , the theoretical framework for evaluating the product knowledge is perfected , and the feasibility of the method is verified by examples .

4 , combining ontology engineering and knowledge modeling field , a ontology - based knowledge modeling principle is established , and a ontology construction method for generalized product design knowledge is proposed , which provides a unified description standard for all product knowledge resources , and a knowledge ontology model framework is constructed for all product knowledge resources .

5 . Based on service - oriented architecture and semantic Web technology , a decision - making task submission and decomposition algorithm is proposed , and the operational flow of call - making collaboration is constructed .
The fuzzy clustering algorithm based on semantics is put forward , and the feasibility of the method is verified by examples .

6 . The principles of system development strategy and function module division and system development tool selection are discussed . The prototype system GPKDSS of decision support system based on generalized product knowledge is developed , and the theory and method presented in this paper are verified by using a special vehicle production enterprise as an example .

【学位授予单位】：西北工业大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：F273.2;TB472

【参考文献】