A method for product appearance design evaluation based on heterogeneous data

The eye-tracking and electroencephalogram data, as physiological information, have been viewed as effective supplements to subjective reporting for guiding the product appearance design. In this context, how to combine heterogeneous information is a challenging question. This study proposes different methods to determine subjective and objective weights of criteria regarding the self-reporting, eye-tracking, and electroencephalogram data for the evaluation of product appearance design. We introduce the probabilistic linguistic term set with interval uncertainty (IUPLTS) to represent complex self-reporting data, and develop a method to aggregate IUPLTSs. An algorithm is proposed to fuse physiological data on the data layer and feature layer. To combine the obtained heterogeneous information, we define an objective weighting method that examines the differences in indicator data and the correlation between indicators, and then use a level difference maximization model to fuse subjective and objective weights. To ensure the stability of decision-making results for the problem involving a large number of indicators, we use the Measurement of Alternatives and Ranking according to the Compromise Solution (MARCOS) to rank alternatives. An example regarding the evaluation of automobile appearance design schemes is presented to show the validity and practicality of the proposed method. The prototype support system of the proposed method has been developed and is freely available at https://github.com/BitSecret/DAQQSO.