【作者】 郭茜；

【导师】 蒲云；

【作者基本信息】 西南交通大学 ， 管理科学与工程， 2013， 博士

【摘要】 质量功能配置是系统化、结构化的产品开发和质量管理方法。该方法采用序贯决策的模式将顾客需求贯穿于应用的全过程,决策信息集中,决策过程透明,规范其中的重要环节,并可自成文档。在当前制造业市场竞争不断加剧、顾客需求的多样化和个性化已成潮流、产品与服务开发管理日益复杂的情况下,研究质量功能配置具有较好的现实意义和应用前景。质量功能配置的过程,是信息流动与处理的过程。从原始顾客需求开始,各方面的相关信息经收集、加工、处理后,形成所需的设计知识,支持后续阶段的决策活动。由于决策对知识使用的特殊需要、知识本身的不清晰性和复杂性、信息呈现形式的多样性和不确定性,质量功能配置在实施过程中存在—定困难。不确定多属性决策旨在解决不确定环境下有限备选方案的多准则决策问题,与质量功能配置的许多问题有内在的紧密联系。为此本文结合不确定多属性决策理论与方法对其中的若干关键问题进行研究,目的是使这种行之有效的产品开发和质量管理方法能够更好地适应不确定的信息环境,丰富和完善质量功能配置的方法体系,为我国制造业企业的产品或产品服务开发提供技术支持,具体的研究成果主要体现在以下方面：1)针对市场调查时顾客评价信息经常存在缺失的实际情况,提出基于顾客群体不完全语义偏好信息的顾客需求基本重要度确定方法。首先充分利用个体的已知偏好信息对缺失的偏好信息进行合理推断,然后通过信息集结获得群体偏好信息,进而揭示顾客视角下的需求的相对重要性。该方法不但解决了实际调查中普遍存在的顾客部分评价信息缺失的问题,而且能够降低评价者的负担,有助于评价者集中于自己所关注的“焦点项目”,从而提高了分析结果的准确度。2)针对产品需求信息的多样性和不确定性,提出一种利用产品规划质量屋中的竞争性评价信息及卡诺模型对顾客需求基本重要度进行调整的方法。为了深入挖掘竞争性评价信息,提出顾客需求的竞争性重要度和改进优先度的概念,分别体现顾客视角下市场竞争的需要和企业从自身实际出发改进产品性能的意愿,推导出二者的计算公式；同时结合卡诺模型的思想,由顾客满意度双因素分析法识别顾客需求的卡诺分类并估计卡诺因子；最后通过上述分析结果对顾客需求基本重要度的修正,对于增强产品市场竞争力、提升顾客满意度潜在贡献较大的需求项目赋予更高的重要度。3)根据质量屋实现顾客需求向技术特征转换时信息来源的特点,提出先分别依据质量屋中的主、客观信息赋权计算再加权集成的技术特征重要度评估方法。进行主观赋权时,允许各专家自由选择语言术语集对顾客需求与技术特征的关系进行评估,并采用语言加权算子和语言混合集成算子进行信息集结,体现数据本身及所在的位置的重要性；进行客观赋权时,允许技术特征值以区间数形式表示,将含有区间数的技术竞争性数据转化为目标相离度矩阵后,使用熵值法赋权。过程中既有专家的经验判断,又有对现实数据的逻辑推理,实现了主观和客观的统一。4)根据产品概念设计方案决策的目标和准则体系的特点,提出基于层次分析法与排序组织法的设计方案决策方法。在建立定性与定量相结合的决策准则体系的基础上,首先使用层次分析法确定属性权重,然后使用排序组织法计算各属性上的偏好指标,进而计算各方案的净流,获得方案的完全序。该方法以设计者的目标与价值观为导向,并兼顾属性信息的不确定性,可为产品设计方案的选择提供一条有效途径。5)利用服务运作和质量功能配置的核心思想相吻合的特点,提出基于质量功能配置的产品服务开发方法体系。在从服务的一般概念和内涵出发对产品服务进行剖析的基础上,使用质量功能配置的方法工具将顾客对产品服务的需求逐步展开为阶段具体目标,设计出适用于产品服务开发的质量功能配置过程,并给出过程中确定服务属性重要度的群决策方法步骤,为产品服务的开发策划创设了一个新的思路。本文在研究中,通过一系列产品和产品服务开发的具体实例验证了各章节所提方法的适用性和有效性。更多还原

【Abstract】 Quality function deployment is a systematic, structured method for product development and quality management to realize customer requirements throughout the whole process of its application in the form of sequential decision. The decision information is centralized and the decision process is transparent. With the normalization of its important processes, the method can be self-documenting. Currently, product market competition continues to accelerate, the diversification and personalization of customer requirements has already become the trend, product/service development is becoming increasingly complex, thus quality function deployment has a good practical significance and application prospects.The process of quality function deployment is the processing flows of information. From the original customer requirements, all the relevant information is collected and handed to refine the required design knowledge to support the subsequent decisions. The vagueness and complexity of knowledge, the diversity and uncertainty of the information form, and the special needs of knowledge for decision making hinder the implementation of quality function deployment. The Uncertain Multiple Attribute Decision Making methods are technical decision aids for evaluating alternatives characterized by multiple attributes under uncertain environment, which has intrinsically close contact with many problems of quality function deployment. This research combines the Uncertain Multiple Attribute Decision Making methods to explore some key issues in quality function deployment. We hope our research can help this effective method better adapt to the uncertainty enviroment, enrich and improve the quality function deployment methodology, and provide technical support for the development of product/service in manufacturing firms. The major achievements and innovations of this dissertation are summarized as follows:1) To deal with the the vagueness and incompleteness of customer preference in customer requirements analysis, a method on the basis of incomplete linguistic judgment matrices to analysis the importance of customer requirements is proposed. Each individual preference is represented by an additive linguistic judgment matrice and extended to a complete form when certain elements are missing. Then all these extended matrices are aggregated into a collective one in order to obtain the importance of customers’requirements. The method can not only solve the problem of information missing in customer surveys, but also reduce the burden of the evaluators, and help them to focus on their own concerns, thus improves the result accuracy.2) To deal with the diversity and uncertainty of information in customer requirements analysis, a method integrating market competitive information in product planning house of quality and Kano model to adjust the fundamental importance ratings of customer requirements is given. In order to min knowledge from the competitive information, the concepts of the competitive priority ratings and the improvement ratios of customer requirements are proposed, which reflect the need for competition from the customers perspective and the firm’s willingness to improve its product performance, respectively. Their solving formulas are also derived. Two-factor analysis is utilized to identify the different categories of customer requirements and estimate the Kano regulators. By using the competitive priority ratings, the improvement ratios and the Kano regulators to correct the fundamental importance ratings, the final importance ratings of customer requirements are determined. It makes designers pay more attention to customer requirements that have larger potential contribution in enhancing the competitiveness and improving customer satisfaction.3) According to the characteristics of information source types in translation from customer requirements to technical characteristics by using house of quality, a synthetic method utilizing the subjective and objective information in house of quality to determine the priority ratings of technical characteristics is proposed. In subjective weighting, decision-makers can choose linguistic terms to evaluate the relationship between customer requirements and technical characteristics according to their own preference. The weighted averaging operator and the linguistic hybrid operator are used to aggregate the individual judgments into a collective value for each technical characteristic, reflecting the importance degrees of both the given argument and its ordered position. In objective weighting, uncertain values of technical characteristics are given in the form of internal numbers. Based on technical competitive information, a goal-deviation matrix is conducted in order to obtain objective importance ratings by using entropy methods. The final result is weighted value of the subjective and objective calculation, which coordinates and unifies the wisdom of experts and the data-based reasoning.4) According to the characteristics of the goal and criteria in decision making for product design alternatives, an evaluation model combining analytic hierarchy process with preference ranking organization method for enrichment evaluations is proposed. Based on the combination of quantitative and qualitative decision criteria system, analytic hierarchy process is used to determine the weight for criteria. Preference ranking organization method for enrichment evaluations is used to calculate the net flow to obtain the ranking of the design alternatives. This approach takes into account the uncertainty of the attribute information and orients to the designer’s goals and values, thus has certain implication in product design.5) Due to the fact that quality function deployment and service operations coincide in their core ideas, a product service deployment methodology based on quality function deployment is proposed. On the basis of analyzing product service from the perspective of the general concept and content of service, quality function deployment tools are utilized to transform customer requirements for product service into the specific objectives of each stage. The process of quality function deployment for product service is designed and a group decision method to determine the importance ratings of service attributes is given, that can creat a new way for product service deployment.Finally, a series of real-world product/service development cases prove the feasibility and effectiveness of the abovementioned methods in this paper.更多还原