Design alternative evaluation based on social utility and market share prediction

Abstract

In new product or service development, design alternatives are generally evaluated by subjective rating, and the one with the highest average score is often chosen. However, the public might not prefer an alternative as much as its average rating suggests. The main objective of this study is to examine the possibility of identifying a predictive function that can accurately represent the products that most people want to purchase in the context of mobile phone products. This paper introduces social welfare functions for developing predictive functions and several metrics to evaluate the functions. Two case studies were performed to determine predictive function for mobile phones. In both case studies, the Atkinson welfare function with a parameter value of 2.5 predicted subjects' purchase decisions and market sales the most. The commonly used average function was not as good as the Atkinson function.