To achieve intelligent design and precise prediction of soil conditioning, this study proposes a hybrid modeling approach integrating machine learning and inverse reasoning. Using sandy soil from the Shenyang Metro Line 3 project as the subject, a dataset containing 120 groups of data was established through laboratory tests, covering key parameters such as effective particle size, foam injection ratio, slurry concentration, and slurry injection ratio. The performance of six machine learning algorithms in predicting soil slump and permeability coefficient was systematically compared. The results show that the Support Vector Regression (SVR) model performed best in both tasks, demonstrating high prediction accuracy and strong robustness. Interpretability analysis based on the SHAP method indicates that effective particle size is the most critical factor influencing conditioning effectiveness, and different output variables exhibit significant differences in sensitivity to input features. To address issues in inverse parameter design such as overfitting and non-unique solutions, this study innovatively proposes an SVR-BPNN hybrid modeling strategy. It employs a BP model to generate parameter combinations and verifies them through the SVR model with iterative optimization, significantly improving the feasibility and reliability of parameter design. The model was successfully applied in the Shenyang Metro Line 3 project. With target slump of 200 mm and permeability coefficient of 2.0×10-5cm/s as inputs, the model inversely derived conditioner parameter combinations that met error requirements. Field measurements aligned well with target values, and shield tunneling proceeded smoothly, validating the model’s effectiveness and practical value in real-world engineering.