GINI index based attribute selection for product configurator design
Editor: Udo Lindemann, Srinivasan V, Yong Se Kim, Sang Won Lee, John Clarkson, Gaetano Cascini
Author: Wang, Yue; Tseng, Mitchell
Institution: Advanced Manufacturing Institute, The Hong Kong University of Science and Technology, Hong Kong S.A.R. (China)
Product configurator system is an important tool to bridge customer needs and company’s offering. It has been widely accepted in industry to facilitate customized product design. However the efficiency of product configuring process has been a challenging issue, especially when the product is complicated. Current configuration systems often perform in a deterministic manner. They cannot adapt to each individual customer preferences by leveraging on the preferences information captured in previous configuration steps. This paper present a Gini index based attribute selection approach for configurator design. Product configuring is modeled as a sequential query-answer process. In each configuring step, Gini index is deployed to quantify the clarity of designer’s belief about the customer’s needs. The attribute which contributes most to the clarity will be select for the customer to configure. As a result, designers get clear about the customer’s needs and preferences in an accelerated manner. An example is presented to test the viability of the method.