Existing microscopic simulation programs allow the users to specify the proportions of heavy vehicles (HV) in the traffic streams and the attributes of these vehicles. Because collecting these attributes for local conditions require extra costs and efforts, most users of simulation models use the default values of the attributes and only change the proportion of the HV relative to passenger cars in the inputs to the simulation programs. However, the impacts of HV on congested freeway sections are highly dependent on the coded HV attributes. Running different simulation programs without calibrating the HV parameters can result in significant differences in the simulated impacts of HV.This study investigates the impacts of HV on congested freeway sections, as assessed by two widely used simulation programs and compares the results to the impacts reported in the Highway Capacity Manual 2000 (HCM 2000) and previous work on the subject. The results confirm the differences between the impacts estimated by different models and methods. It is recommended that the simulation models are calibrated to achieve acceptable passenger car equivalency values of HV. This paper also shows that the default HV mix of one of the widely used simulation programs includes truck types that are unable to handle step grades in the simulation. Modifications to the default values in the program and/or examination of the program modeling of HV are recommended to address this issue.
Abstract