The reliability of capacity of signalized intersections in mixed traffic conditions involving vehicles, bicycles, and pedestrians was investigated to complete the conventional, deterministic capacity calculations. Simulations using VISSIM provided estimates of capacity distributions, and demonstrated the effects of the analysis intervals on the distributions. With the random vehicle arrivals taken into account, a capacity reliability assessment method was given as a function. Assessments were also performed regarding the effects of the conflicting pedestrian and bicycle volumes on capacity reliability. The simulation indicates that the pedestrians and bicycles result in greater random fluctuations of exclusive turning lane capacities, but have less effect on the variability of shared lane capacities. Normal distributions can be used to model the capacities for intervals not less than 10 min. At higher vehicular volumes, the capacity reliability is more sensitive to the mean and standard deviation of the pedestrian and bicycle volumes.

Bicycle traffic has a significant effect on the capacity of signalized intersections. This paper divides the influence of bicyclists on vehicular flow into four types with the time durations estimated based on probability, shock wave, and gap acceptance theory. Vehicular saturation flow rate is predicted for various conditions on the basis of the speed-flow curve for the capacity of typical intersections influenced by bicycle traffic. The model overcomes the limitations of the Highway Capacity Manual (HCM, 2000) method for left-turns due to data collection, and takes into account the effect of trapped bicycles on the through vehicular traffic. The numerical results show that the left-turn and through capacities predicted by the model are lower than those of the HCM method. The right-turn capacity is close to that of the HCM method at low bicycle volumes and higher than that of the HCM method at high bicycle volumes.