Self-compacting concrete (SCC), owing to its ability to flow and consolidate under its own weight, is considered one of the strategic materials for the construction of complex structures. The simultaneous dependence of its mechanical performance on the type, content, and combination of fibers has transformed mix design into a multivariable problem, making the application of intelligent data-driven approaches essential. The present study was conducted to intelligently optimize the mix design of self-compacting concrete reinforced with Forta, glass, and ceramic fibers and to develop a data-driven decision-support framework using the Charged System Search (CSS) algorithm. Data obtained from 274 experimental specimens cured for 7 to 180 days were analyzed, and multiple regression models were developed as surrogate models to predict compressive, tensile, and flexural strengths. The results showed that Forta fibers had the greatest positive effect on improving the mechanical performance of concrete compared with glass and ceramic fibers. Specifically, the mixture containing 5% Forta fibers achieved a compressive strength of 23.2 MPa, a tensile strength of 6.12 MPa, and a flexural strength of 10.30 MPa after 180 days of curing. Furthermore, in the single-objective optimization, the CSS algorithm proposed a mix design consisting of 0.73% Forta fibers, 0.88% glass fibers, 0.04% ceramic fibers, and a water-to-cement ratio of 0.387, which increased the compressive strength from 15.78 MPa to 22.29 MPa, representing a 41.3% improvement. In the multi-objective optimization, the tensile strength increased by 23.6%; however, this improvement was accompanied by a reduction in compressive strength, revealing the existence of a trade-off among the mechanical objectives. The R² values for the compressive, tensile, and flexural strength models were 0.514, 0.370, and 0.243, respectively, indicating that the model achieved higher predictive accuracy for compressive strength than for the other responses. The findings demonstrate that integrating data analysis with the Charged System Search algorithm provides an intelligent decision-support framework for screening the design space, selecting optimal mix designs, reducing the number of experimental tests, and supporting decision-making in the design process of self-compacting concrete. Nevertheless, independent experimental validation of the proposed optimal mix designs is essential before their implementation in practical applications.