Steel fiber (SF) and macro-polypropylene fiber (MPPF) were used as aggregate to prepare hybrid fiber recycled aggregate concrete (HFRAC) suitable for pavement in order to improve the strength and wear resistance of recycled aggregate concrete (RAC). The Box-Behnken experimental design method was adopted, with the volume content of SF, volume content of MPPF and sand ratio as factors, and the flexural strength, compressive strength and wear amount of HFRAC as evaluation indicators. The prediction model of each evaluation index was established, the influence of each factor on evaluation index was analyzed, and the optimal model of the mix proportion of HFRAC based on NSGA-Ⅱ coupled entropy weight TOPSIS method was constructed to determine the mix proportion scheme when the comprehensive performance was optimal. The results show that the response surface model of each evaluation index has good fitting effect and high prediction accuracy within the test range. Each evaluation index is not only affected by a single factor, but also affected by the interaction between factors. The interaction between the volume content of SF and the volume content of MPPF has a extremely significant effect on flexural strength and compressive strength, and has a significant effect on wear amount. The interaction between the volume content of MPPF and the sand ratio has a significant effect on flexural strength. The optimal mix proportion of HFRAC is obtained. When the volume content of SF is 1.39%, the volume content of MPPF is 0.97%, and the sand ratio is 36.10%, the comprehensive performance of HFRAC is the best. This method can realize the comprehensive improvement of HFRAC performance, and provide a certain theoretical basis and technical support for the popularization and application of HFRAC in road engineering.