SUMMARY

Recent advances in social media that allow better accesses to social networks of customers have profound technical and economic implications for design and innovation research. This research is motivated to investigate social network effects on customer preference and product demand modeling for design with a focus on the interface of engineering, marketing, and social computing. Designing a product that a customer likes and intends to purchase requires not only an in-depth knowledge of how engineering performance is linked to customers desired product attributes, but also a profound understanding of how they make product choices in a broad context, including risks, affective motivations, cognitive tendencies, and social network effects, in which a customer’s choice of a product is (substantially) influenced by the rest of his or her social network. However, traditional efforts in product demand modeling are primarily based on customer preference, such as utility theory and discrete choice models, which unfortunately neglects the fact that a customer’s strong preference of a product may not necessarily be translated into his purchase decision of this product. In order to predict product demand, the key question is how to predict customer’s product adoption (purchase) probabilities. This research aims to extend customer preferences and choice modeling to prediction of product adoption probability, while incorporating social network effects and accounting for decision makers’ subjective experiences (e.g., affective responses and cognitive tendencies) and risk attitudes. From the perspective of engineering design, we will explore product innovation to mine customer preferred product attributes through trend analysis and knowledge discovery from social media. From the decision making point of view, we will investigate the influence of psychological factors (e.g., risks, affective elements, cognitive tendencies) on customer preference modeling. From the marketing point of view, we will predict customer’s product adoption probabilities, incorporating customer’s hurdle prospect and social network effects.