With the development of intelligent connected vehicle technology, roads will exhibit driving characteristics of mixed heterogeneous traffic flow. Due to the significant differences in vehicle performance and driving characteristics between intelligent connected vehicles and traditional manually driven vehicles, road traffic capacity will change. This study first analyzed the factors affecting road capacity under mixed heterogeneous traffic flow, with a focus on factors such as workshop time distance, CAV vehicle penetration rate, and vehicle mixing type; Secondly, based on the Wiedemann car following model and analysis data from PATH laboratory, a vehicle car following model for mixed heterogeneous traffic flows was constructed; Finally, a test section was selected to simulate and calculate the road traffic capacity under different CAV vehicle penetration rates using VISSIM simulation software. The results showed that compared to traditional manually driven vehicles, when the CAV vehicle penetration rate reached 100%, the intersection traffic capacity would increase by 50% and the section traffic capacity would increase by 40%.