Comparison of Estimates of Travel Time Losses on High-Capacity Roads

Abstract

This paper compares two approaches to estimating travel time losses on freeways that are based on a whole-year analysis. Both approaches are intended to improve the design process for freeways by accounting for the annual traffic demand profile as well as short time variance of demand and capacity. One approach estimates breakdown probability and delayrelated time losses by using a probabilistic model. The other approach uses Monte Carlo simulation to estimate travel time losses at freeway bottlenecks on the basis of a queuing model with stochastic demand and capacity. Both approaches are based on annual hourly demand profiles. The approaches are compared for demand profiles and deliver similar but not equal results. Travel time losses for both approaches increase exponentially with traffic volume, with the simulation-based approach consistently generating higher time losses for higher traffic volumes. Both approaches are a substantial improvement on current design procedures, are easy to apply, and are recommended for adoption in traffic analysis guidelines.