<p>Pervious concrete (PC) is increasingly adopted in sustainable pavement systems owing to its ability to reduce stormwater runoff, mitigate flooding, and support urban heat management. Despite these advantages, designing mixtures that balance permeability and strength remains challenging because improving one property often reduces the other property. Conventional trial-and-error approaches are inefficient, underscoring the need for systematic methods that can simultaneously optimize multiple performance objectives at the same time. In this study, a combined framework of Response Surface Methodology (RSM), Sobol sensitivity analysis, and symbolic regression was used to investigate the influence of aggregate gradation, cement-to-aggregate ratio (C/A), and water-to-cement ratio (W/C) on the hydrological and mechanical properties of PC. Experimental testing of the designed mixtures enabled the development of predictive models, ranking of factor importance, and discovery of interpretable equations linking the mixing parameters to performance. The results showed that the W/C ratio predominantly governs the porosity, the C/A ratio controls the permeability, and the aggregate gradation strongly affects the compressive strength. Multi-objective optimization identified a mixture that balanced permeability and strength, and validation confirmed its accuracy. The integrated methodology offers a structured approach to PC mix design, reduces the experimental effort, and supports practical decision-making. These findings provide tools for developing efficient and sustainable pavement solutions, particularly in resource-constrained urban contexts.</p>