Product engineering characteristics dynamic evaluation and optimization based on Bayesian network

In order to describe the uncertain relationship between evaluation indicators of product and evaluate the product solution dynamically and rapidly, a dynamic evaluation method based on Bayesian network is proposed. First, we establish and analyze the evaluation indicators of design solutions, and build a Bayesian network model; second, the parameters of the Bayesian network are calculated by expert experience or maximum likelihood estimation in different application environments; then the indicators that have the highest influence on product evaluation are computed based on the sensitivity analysis in order to obtain the improvement direction of product design. Finally, the proposed method is verified by the example of a subway vehicle traction motor.