This study focuses on the development of a predictive model for the compressive strength of concrete using recycled brick aggregate with the CatBoost algorithm (CAT) and SHAP value analysis to investigate the influence of input variables on the prediction results. The Sparrow Search Algorithm (SSA) is also utilized to optimize the CAT model and find the best-performing one. The data used in the study were collected from the relevant literature, including 11 input variables and 1 output, with a total of 393 experimental samples. During the development phase, the dataset was divided into training and testing sets in a 70:30 ratio. The results showed that the predictive model could accurately predict the compressive strength of recycled concrete, and variables such as cement content, concrete age, total coarse aggregate amount, and percentage of coarse crushed brick aggregate significantly influenced the prediction results.