An integrated framework for predicting consumer choice through modeling of preference and product use data

A critical task in product design is mapping information from consumer space to design space. Currently, this process is largely dependent on the designer to identify and map how psychological and consumer level factors relate to engineered product attributes. In this way, current methodologies lack provision to test a designer's cognitive reasoning, which could introduce bias while mapping from consumer to design space. Cyber-Empathic Design is a novel framework where user-product interaction data is acquired using embedded sensors. To understand consumer perceptions about a particular product, a network of latent psychological constructs is used to form a causal model allowing designers to better understand user preferences. In this work, we extend this framework by integrating choice-based preference modeling to develop a Discrete Choice Analysis integrated Cyber-Empathic design framework (DCACED). We model user preferences and ultimately consumer choice by considering perceptions estimated by psychological latent variables and user-product interaction data. To demonstrate the effectiveness of the framework, a case study using a sensor integrated shoe design is presented where data to represent user demographics, sensor readings, and product choice is simulated. Using the DCA-CED method, the model parameters are recovered and compared with the original parameter values in the simulator. In addition, the ability of the framework to predict choice based on user product-interaction data is tested. The results show that the analytical method effectively captures the underlying data generation process thereby validating the proposed framework and the analytical method.