A Multidimensional Network Approach For Modeling Customer-Product Relations In Engineering Design

Mingxian Wang,Yun Huang,Wei Chen,Noshir S. Contractor,Yan Fu

DOI: https://doi.org/10.1115/detc2015-46764

2016-01-01

Abstract:Analytical modeling of customer preferences in product design is inherently difficult as it faces challenges in modeling heterogeneous human behavior and product offerings. In this paper, the customer-product interactions are viewed as a complex socio-technical system and analyzed using social network theory and techniques. We propose a Multidimensional Customer-Product Network (MCPN) framework, where separate networks of "customers" and "products" are simultaneously modeled, and multiple types of relations, such as consideration and purchase, product associations, and customer social networks are considered. We start with the simplest unimodal network configuration where customer cross-shopping behaviors and product similarities are analyzed to inform designers about the implied product competition, market segmentation, and product positions in the market. We then progressively extend the network to a multidimensional structure that integrates customer preference decisions with product feature similarities to enable the modeling of preference heterogeneity, product association and decision dependency. Finally, social influences on new product adoption are, analyzed in the same framework by introducing customer-customer relations together with other product-product and customer product relations. Beyond the traditional network descriptive analysis, we employ the Exponential Random Graph Model (ERGM) as a unified statistical inference framework for analyzing multiple relations in MCPN to support engineering design decisions. Our approach broadens the traditional utility based logit approaches by considering the dependency among product choices and the "irrationality" of customer behavior induced by social influence. While this paper is focused on presenting the conceptual framework of the proposed methodology, examples on customer vehicle preferences are presented to illustrate the progressive development of the MCPN framework from a simple unimodal configuration to a complex multidimensional structure.