Integration of customer and designer preferences in robust design

Robust design is one of the most effective design methodologies for minimising the variability of product performance, thereby improving the quality of a product. Although robust design principles have been widely implemented in many industries and a number of new robust design models have been reported in the literature of Design for Six Sigma, there is still ample room for improvement. Firstly, most robust design models consider a single quality characteristic. In reality, however, the judgmental basis of a product from the perspective of the customer is characterised by the assessment of multiple characteristics. Secondly, previous models may have ignored customer and designer preferences in modelling and optimisation phases. To rectify these shortcomings, this paper proposes a systematic weight assessment method to establish the combined preference structure embedding customer and designer preferences. Finally, this paper proposes a model for synthesising Pareto solutions using the weighted Tchebycheff metric that represents compromised trade-offs between multiple quality characteristics based on a combined preference structure.