Research Of Quality Function Deployment In Uncertain Information Environment

In 1994, the famous quality management master, Dr. Juran, said that the 21st century was the century of quality. In the century of quality, with the increasing intensity of competition and establishment of buyer market, customer satisfaction has becoming one of the most important driven forces to succeed in the market. As one of the upstream quality management methodology, quality function deployment (QFD) occupies very important position in the entire quality management disciplines, which is also a very powerful weapon for enterprises to achieve customer satisfaction. A good many QFD application cases show that QFD can effectively shorten product development cycle, enhance internal communications, improve product/service quality, reduce project and product cost, increase customer satisfaction, and finally, improve enterprise performance.Generally, QFD was used in the early stage of new or improved products/services development process. With the increasing complexity of social-economic system, QFD team members usually do not have sufficient knowledge and information needed, and therefore most of their judgments and evaluations are highly subjective, imprecise and uncertain in nature, which will significantly impact the quality of the QFD results. In view of this, a method or approach that is capable to systematically analyze and accurately quantify those subjective evaluations and judgments of the QFD team member is highly required. Therefore, aiming at the typical uncertain information problems in product development, deep and systematic research is done on both the theory and application aspects of QFD in this dissertation. The major achievements and innovations of the dissertation are summarized as follows:1) To deal with the uncertain group decision-making (UGDM) problems in customer requirements (CRs) analysis in QFD, some new methods for determining CRs importance ratings were developed. Firstly, a rough Analytic Hierarchy Process (RAHP) approach to obtain the importance rating of CRs in QFD was proposed. RAHP was based on the integration of the AHP and two novel concepts known as rough numbers and rough boundary intervals, which were derived from the basic notions of rough sets. Secondly, the decision-making theory with linguistic information developed in recent years was introduced and extended, and some transformation functions among multi-granular unbalanced linguistic labels sets were defined. Then, based on the relative ideal solutions and linguistic judgment matrixes, two novel methods based on multi-granular and mixed types'linguistic information were proposed to obtain the fundamental importance of CRs in QFD. Lastly, market competition information was deeply investigated in this study, and a method for determining market competitive priority ratings of CRs based on uncertain linguistic variables was proposed. The final importance ratings of CRs were therefore obtained by the integration of the fundamental importance of CRs, the market competitive priority ratings of CRs, and the priority ratings for achieving the improvement targets of CRs'satisfaction degrees.2) To deal with the UGDM problems in technical characteristics (TCs) analysis in QFD, a new linguistic information aggregation operator, dependent linguistic ordered weighted averaging (DLOWA) operator, was firstly developed. The CRs-TCs functional relationships presented with linguistic judgments were aggregated by the DLOWA operator, which could relieve the influence of unfair linguistic judgments. Afterwards, a method for determining the fundamental priority ratings of TCs under multi-granular linguistic information environment was presented. Secondly, technical competition information was deeply investigated, and a method for determining the technical competitive priority ratings of TCs based on uncertain linguistic variables was proposed. Thirdly, the efficiency and risk analysis were implemented on the achievement of the improvement target of TCs'competitive performance, an algorithm to calculate the priority ratings for achieving the improvement target of TCs'competitive performance was presented. Lastly, the final priority ratings of TCs were determined by the integration of the fundamental priority ratings of TCs, the technical competitive priority ratings of TCs, and the priority ratings for achieving the improvement targets of TCs'competitive performance.3) Aiming at dynamic uncertainty of CRs in product development process, a dynamic QFD analysis method based on the improved grey forcasting model was developed. Firstly, a new optimization model based on the combination of the initial condition and the background value was proposed to improve the accuracy of the traditional grey forecasting model. Then, some simulation data were used to compare and verify the precisions of the improved grey forecasting model. It was showed that the improved grey forecasting model was more precise and adaptive. Thirdly, a dynamic house of quality (DHOQ) framework was introduced, and the improved grey forecasting model was integrated into the DHOQ framework to fit and forecast dynamic customer requirements. Therefore, the trend of the importance of each technical characteristic could be monitored, analyzed and controlled using the DHOQ so as to fulfill dynamic customer requirements and allocate the limit design resources in a better way.4) To solve technical contradiction issues in product design and development process, a method for solving technical contradictions in QFD process based on the Russian Theory of Inventive Problem Solving (TRIZ) was proposed. Firstly, QFD methodology was used to systematically identify the technical contradictions in product development. Specifically, the technical characteristic pairs of which relationship were negative in the technical correlation matrix could be found out and recognized as the technical contradictions. Secondly, the inventive principles of TRIZ were applied to solve these technical contradictions, and some feasible innovative solutions or inspirations could be acquired. These innovative solutions should be evaluated by product development experts and the best feasible one would be chosen. Considering the vagueness and uncertainty of expert judgment, and the case of multi-granularity and multi-semantics in an expert's linguistic statement, the linguistic information decision-making theory was introduced. Lastly, two new methods, which were the method for product innovation scheme evaluation based on multi-granular linguistic information and the method for product innovation scheme evaluation based on free mixed linguistic information, separately, were proposed to evaluate and select the best feasible alternative under uncertain information environment.Finally, the applicability, exactness and feasibility of the abovementioned methods were illustrated by a series of real-world product development cases. Theoretically, the above advancement in QFD methodology enriches and develops QFD theory. Practically, the above advancement in QFD as well as the application cases will be of great importance that points out struggling orientation for companies to implement QFD method successfully.