In this paper, it studies the various technical indicator values of 12 women's basketball teams in the 30th Olympic Games, in the research process uses neural network-based fuzzy comprehensive evaluation method. First it expounds the principle and method of the single-level fuzzy comprehensive evaluation model, multi-level fuzzy comprehensive evaluation model and the evaluation results test, and then focuses on the construction method of neural network structure and the determination method of the connection weights in neural network method, and finally on the basis of the theory conducts reasonable comprehensive evaluation on each team's offensive ability, defensive ability and comprehensive offense and defense abilities for the 30th session’s technical indicators data. Through the analysis of the results: the comprehensive evaluation results are in consistent with the actual competition rankings, this method has a higher feasibility in the process of applying, and it provides feasible suggestion for the team's future development direction.