Optimization of quality function deployment process: A decision support system based on multi-objective programming approach

The purpose of this thesis is to optimize Quality Function Deployment (QFD) process for discrete values of design requirements (DRs) in new product design or development of an existing product. In the QFD literature, for optimizing product development, determining the target levels of DRs that are continuous range is crucial. However, in practice, values of DRs are often discrete. Therefore, unlike the existing studies, a new Mixed Integer Linear Programming-1 (MILP-1) model is proposed by considering that values of DRs are discrete for QFD optimization. In this model, Kano model is used in order to integrate customer satisfaction levels into QFD optimization process. In addition, same as the existing literature, it is assumed that there is a linear relationship between fulfillment levels of DRs and their costs. But, such a relationship in real cases rarely emerges due to the technological and production constraints. Besides, QFD process may necessitate to simultaneously optimize more than one conflicting objectives to obtain more realistic solutions. It is really hard for decision makers to determine the target values of these objectives in imprecise and uncertain environment. Because of these reasons, Mixed Integer Linear Programming-2 (MILP-2), Mixed Integer Goal Programming (MIGP) and Fuzzy Mixed Integer Goal Programming (FMIGP) are developed respectively by extending the MILP-1 model. Finally, a Decision Support System (DSS) is developed in order to assist decision makers to effectively use the proposed models and make product design easily and faster. Applicability of developed DSS called QFD Analist is illustrated in a real life problem.