Fuzzy regression-based mathematical programming model for quality function deployment

Abstract

Quality function deployment (QFD) is becoming a widely used customer-driven approach and tool in product design. The inherent fuzziness in QFD modelling makes fuzzy regression more appealing than classical statistical tools. A new fuzzy regression-based mathematical programming approach for QFD product planning is presented. First, fuzzy regression theories with symmetric and non-symmetric triangular fuzzy coefficients are discussed to identify the relational functions between engineering characteristics and customer requirements and among engineering characteristics. By embedding the relational functions obtained by fuzzy regression, a mathematical programming model is developed to determine targets of engineering characteristics, taking into consideration the fuzziness, financial factors and customer expectations among the competitors in product development process. The proposed modelling approach can help design team assess relational functions in QFD effectively and reconcile tradeoffs among the various degree of customer satisfaction and determine a set of the level of attainment of engineering characteristics for the new/improved product towards a higher customer expectation within design budget. The comparison results under symmetric and non-symmetric cases and the simulation analysis are made when the approach is applied to a quality improvement problem for an emulsification dynamite packing machine.

Acknowledgements

The paper was joint financially supported by the National Natural Science Foundation of China (NSFC 70002009), the Excellent Youth Teacher Program of Ministry of Education of China and the Shenyang Natural Science Foundation (1020036-1-03) and partly by a Strategic Research Grant (SRG) from City University of Hong Kong (project no. 7001227). The authors are indebted to the Editor and referees for invaluable comments and suggestions on the paper.