Prediction of minimum miscibility pressure (MMP) of the crude oil-CO2 systems within a unified and consistent machine learning framework

•A 1D-CNN model is developed for predicting MMP between CO2 crude oil within a unified and consistent framework.•MMPs measured by using both slim-tube and RBA methods are ddifferentiated for the first time with a marked parameter.•The SHapley Additive exPlanations (SHAP) was introduced for the first time to interpret the proposed model embedded with correct physics.•Taking single component as the input for forecasting the MMP greatly improves the prediction accuracy compared with multi-pseudocomponents.•MMPRBA is generally larger than MMPST, and the effect of each influential factor on the respective MMP is consistent.