The merging area is a typical traffic congestion point and accident-prone zone. As automated vehicles (AVs) improve and become more commercially available, they will coexist with human-driven vehicles (HDVs) for a long time. However, current methods struggle to provide convincing evaluation results for different AVs penetration rates in merging areas. To address this, a mixed traffic simulation framework was designed to evaluate the impact of varying AVs penetration rates on traffic safety and efficiency in merging areas. This framework incorporates the intelligent driver model (IDM) to describe HDVs longitudinal behavior and a mandatory lane-changing model to characterize HDVs lane-changing behavior under pressure of different positions. For AVs, adaptive cruise control (ACC) and cooperative adaptive cruise control (CACC) models were used to simulate car-following behavior. Polynomial curve-based models were employed to plan AVs’ lane-changing trajectories. A merging zone scenario with different flow rates and AVs penetration rates was developed using SUMO. Results indicate that AVs can significantly improve traffic safety and efficiency in merging areas.