In light of increasing resource constraints and environmental concerns, optimizing the cost and performance of high-strength concrete (HSC) has become a key objective. This study introduces a cost-based optimization approach for HSC mix design using experimental data and nonlinear regression models. A dataset comprising 36 HSC mixes across three strength levels (50, 60, and 70 MPa) was used to develop predictive equations for compressive strength and slump. These models were integrated into a Sequential Quadratic Programming (SQP) framework to identify optimal mix proportions. Validation through laboratory tests confirmed the model’s reliability, with optimal designs achieving a water-to-cement ratio of 0.3 and 10% silica fume content. The proposed method reduces material costs while meeting performance criteria and facilitating automated mix design processes
