<p>This research aims to address the persistent traffic conflicts and social dilemmas arising from mandatory lane changing behaviors for bus exiting at bus bay stops in mixed traffic environments. An evolutionary game theory framework is developed, incorporating the influence of different neighborhood sizes on players’ interactions. A dynamic cooperation mechanism is further designed to adjust players’ utilities in real-time and motivate cooperative behaviors between buses and passenger cars. Numerical simulations reveal that the evolutionary game predominantly evolves toward an asymmetric equilibrium where one player cooperates and the other defects. Increasing the neighborhood size accelerates convergence and amplifies strategy trends, leading to more effective conflict suppression. The implementation of the dynamic cooperation mechanism further significantly reduces collision risks. Notably, larger neighborhood sizes within the mechanism facilitate conflict elimination and convergence efficiency. The proposed approaches effectively mitigate the conflicts and improve the safety and operational efficiency of mandatory lane changing behaviors. The findings offer valuable theoretical insights and practical guidance for optimizing behavioral decision-making in public transportation systems.</p>