In response to the discrepancy between the mechanical-electric torque distribution rules used in the shifting calculation of plug-in hybrid electric vehicles and the actual operating conditions, this study focuses on a specific vehicle model known as the multi-gear and multi-mode parallel plug-in hybrid electric vehicle (MGMMP-PHEV). Various control parameters including vehicle speed, pedal opening, battery state of charge (SOC), and operating mode are taken into account by considering the impact of coordinated operation between the dual power sources and battery SOC. In order to address this issue, an enhanced dual-parameter shifting calculation principle and method are proposed, and an optimized shifting strategy is developed with the objective of achieving the highest system comprehensive efficiency and ensuring battery charge-discharge balance. The shifting points under each mode are optimized offline using the DIviding RECTangles (DIRECT) algorithm, and a multi-parameter fuzzy controller is introduced to dynamically adjust the shifting speed. The improved shifting strategy, implemented in Matlab/Simulink, is compared with the initial shifting strategy in terms of its economic performance. The results demonstrate that the DIRECT-optimized and dynamically adjusted fuzzy shifting strategy effectively maintains the balance of battery charge and discharge. Specifically, under the WLTC working condition, it achieves a 5.59 % reduction in fuel consumption compared to the dual-parameter shifting strategy. Furthermore, this optimized strategy leads to a slight increase in shifting delay and a decrease in shifting frequency, thus verifying the effectiveness and superiority of the proposed controller.