In this work, shale oil hydrogenation was upgraded in a two fixed-bed reactors to maximize light fraction with ultra-low sulfur and nitrogen. Three types of catalyst grading for shale oil hydroprotection, hydrofining, and hydrocracking were investigated to evaluate the hydrogenation performance. A lumping kinetic model of hydrodesulfurization, hydrodenitrification, and hydrocracking based on catalysts grading was established for the first time. Model parameters were determined using the Levenberg-Marquardt optimization algorithm. Results revealed that the model could accurately predict the removal rations of sulfur, nitrogen, and desired product yields in shale oil hydrogenation under different catalyst grading. Thus, the model can be used for optimization of catalyst stacking to meet the requirements of hydrogenation products. (C) 2015 Elsevier Ltd. All rights reserved.