Product family architecture design with predictive, data-driven product family design method

This article addresses the challenge of determining optimal product family architectures with customer preference data. The proposed model, predictive data-driven product family design (PDPFD), expands clusteringbased approaches to incorporate a market-driven approach. The market-driven approach provides a profit model in the near future to determine the optimal position and number of product architectures among product architecture candidates generated by the k-means clustering algorithm. An extended market value prediction method is proposed to capture the trend of customer preferences and uncertainties in predictive modeling. A universal electric motors design example is used to demonstrate the implementation of the proposed framework in a hypothetical market. Finally, the comparative study with synthetic data shows that the PDPFD algorithm maximizes the expected profit, while clustering-based models do not consider market so that less profit can be achieved.