Road Network Capacity Reliability Considering Travel Time Reliability

In order to investigate the interaction between travel time reliability and road network capacity reliability, a bi-level programming model based on travel time reliability is set up for the evaluation of capacity reliability in this paper. In the model, the object is to maximize the basic OD traffic demand multiplier in the upper level. For this purpose stochastic user equilibrium traffic assignment is used to describe travelers’ route choice behaviors in the lower programming level, and a prescribed travel time threshold is set in the upper level problem as a constraint to travel time. By assuming that link capacity is continuous truncated normal random variable, using Monte Carlo simulation technique in conjunction with the sensitivity analysis method of road network equilibrium flow, a heuristic algorithm is established to estimate road network capacity reliability. Numerical study on a small road network is presented to demonstrate the validity of the proposed model and algorithm. At the same time, the impact of the travel time threshold, the traffic demand threshold, the traveler's perception error, the variance of the link capacity and the level of OD traffic demand on the road network capacity reliability are fully examined.