Aiming at the uncertainty in the results of the scheming optimization process influenced by subjective factors and the problem of optimal scheme selection during asphalt concrete mix designing, taking the mix design of the Qiaolabula asphalt concrete core wall project as an example, the projection pursuit regression(PPR) modelling technique combined with the TOPSIS comprehensive evaluation method based on entropy weights (PPR-TOPSIS analysis method) was proposed for the comprehensive benefits of the mix evaluation indexes. The results show that the established PPR model for asphalt concrete evaluation indexes is highly accurate in calculation, with a pass rate of 93.1% and an average relative error of 2.01%, which can accurately predict the evaluation indexes of the mix ratio schemes with different combinations of factor levels. The test results of the best mix ratio scheme analyzed by the PPR-TOPSIS comprehensive evaluation method can satisfy the specification criteria. Compared with the traditional design method, the influence of human subjective factors can be reduced and the solution search and quantitative evaluation of the combination of all test factors can be realized by the proposed method, and the obtained ratio solution is more reasonable.