Abstract: The continental mud shale has a strong heterogeneity, and the prediction of its mineral composition, TOC (total organic carbon content) and other parameters is critical for the evaluation and optimization of continental shale oil and gas abundance. To obtain the vertical distribution characteristics of these key parameters of the continental shale formation, and to ovoid the practical problems such as time-consuming experiments, high cost, and sampling difficulty etc., a research of the logging prediction method for key parameters was carried out. The lower sub-member of the third member of the Eocene Shahejie Formation (Es₃) in the Dongying Sag of the Bohai Bay Basin was employed as an example, XRD measurements and logging were carried out based on which, various TOC prediction models were selected for different lithologies, and some TOC identification templates of different lithologies were established.The constructed model was then used to perform typical predictions on the lithology and TOC of the lower sub-member of the Shahejie Formation in well Fy1.The correlation coefficient between the predicted values obtained by the segmented prediction method for TOC and the measured values appeared to be relatively high (about 0.84), indicating that the method developed here is highly feasible and can provide effective technical support for the evaluation and target optimization of continental shale oil resources.