Computationally intensive tasks often require substantial energy support, and mobile devices may experience slow task execution, response delays, and device overheating due to insufficient processing capabilities. By offloading computationally intensive tasks, the energy consumption of mobile devices can be reduced, and the resource utilization and network balance of mobile communication networks can be improved. Therefore, this paper designs a refactored priority division method for offloading computationally intensive tasks in mobile communication networks. This method first calculates the execution delay and energy consumption of mobile communication network system devices, mobile edge computing servers, and cloud servers, and uses these as factors for the refactored priority division of computationally intensive tasks. Then, it employs the Analytic Hierarchy Process (AHP) to refactor the priority division of computationally intensive tasks in mobile communication networks. In addition, deep reinforcement learning is used to comprehensively consider multiple factors such as delay and energy consumption to find the optimal offloading decision. Experimental results show that this method can effectively offload computationally intensive tasks from mobile devices and mobile edge computing servers and improve the balance of mobile communication networks.