Back to results list
Show full item record
Please use this identifier to cite or link to this item:
|Title:||Modelling of customer satisfaction and determination of specifications for new products using computational intelligence techniques||Authors:||Jiang, Huimin||Degree:||Ph.D.||Issue Date:||2014||Abstract:||Affective design and the determination of engineering specifications are two important processes in the early product design stage, especially for consumer products, and the two processes are commonly conducted separately. Generally, designers and engineers are required to determine the settings of design attributes (for affective design) and engineering requirements (for engineering design) respectively for new products. Some design attributes and some engineering requirements could be common, however, the settings of the design attributes and engineering requirements could be different because of the separation of the two processes as mentioned before. In previous studies, a methodology or a framework that considers the determination of the settings of the design attributes and engineering requirements simultaneously was not found. To bridge this gap, a methodology for considering affective design and the determination of engineering specifications of a new product simultaneously is proposed in this research by which optimal settings of the engineering requirements and design attributes can be determined for maximizing customer satisfaction. The proposed methodology mainly involves five processes: conducting of survey, modelling of customer satisfaction based on quality function deployment (QFD), modelling of affective relationships, formulation of an optimization model, and determination of optimal settings of engineering requirements and design attributes. Customer satisfaction models are developed to relate customer satisfaction to design attributes and engineering requirements. However, explicit customer satisfaction models that can capture both the fuzziness and nonlinear behaviour of the modelling are difficult to develop, and deficiencies of the approaches used in previous studies can be seen. In this research, novel approaches to developing customer satisfaction models are proposed.
To develop customer satisfaction models based on QFD, a chaos-based fuzzy regression (FR) approach, is proposed which employs an FR method to determine the fuzzy coefficients of the models, and a chaos optimization algorithm to generate the nonlinear polynomial structures of the models. Regarding the modelling of affective relationships, an adaptive neural fuzzy inference system (ANFIS) was introduced to model the affective relationships and was shown to be an effective approach for the modelling. However, ANFIS was found to be incapable of modelling problems that involve a number of inputs. In this research, rough set and particle swarm optimization (PSO)-based ANFIS approaches are proposed to model the affective relationships in order to make up the deficiency and further improve the modelling accuracy. To determine the optimal settings of the design attributes and engineering requirements for maximizing customer satisfaction, an optimization model is developed that involves the two types of customer satisfaction models. By solving the optimization model using a chaos-based non-dominated sorting genetic algorithm- (NSGA-), the optimal settings of the engineering requirements and design attributes of a new product can be determined. A case study of a mobile phone design was conducted to illustrate the proposed methodology. Validation tests were conducted and some important results were obtained. First, the proposed chaos-based FR approach to modelling customer satisfaction based on QFD outperforms those approaches previously used in the same modelling, including statistical regression, fuzzy regression and fuzzy least-squares regression approaches, in terms of mean relative errors and variance of errors. Second, the proposed rough set and PSO-based ANFIS approaches are able to model affective relationships that involve a number of inputs, and the proposed approaches outperform fuzzy least-squares regression, fuzzy regression and genetic programming based fuzzy regression approaches in modelling affective relationships in terms of mean relative errors and variance of errors. Third, the proposed chaos-based NSGA- for determining the optimal solutions was found to be better than NSGA- in terms of customer satisfaction and diversity of Pareto solutions. Finally, the customer satisfaction values obtained based on the proposed methodology was found to be higher than those obtained based on the combined standalone QFD and standalone affective design approach.
Product management -- Data processing
Hong Kong Polytechnic University -- Dissertations
|Pages:||xxii, 183 leaves : ill. (some col.) ; 30 cm.|
|Appears in Collections:||Thesis|
View full-text via https://theses.lib.polyu.edu.hk/handle/200/7473
Citations as of Jun 4, 2023
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.