New models to evaluate the level of service and capacity for rural multi-lane highways in Egypt

Abstract

Multi-lane highways represent the majority of the total length of highway network in Egypt. The road geometry and the percentage of heavy vehicles (HVs) are considered the most important factors affecting the level of service (LOS) and capacity for any roadway. Therefore, this paper aims to explore the relationship between the road geometric characteristics and HV, and the LOS and capacity by two ways. First is the statistical modeling and second is the modeling by artificial neural networks (ANNs). In this research, the traffic and road geometric data are collected from mid-tangent points at 45 different sites that are located in desert and agricultural highways. The results showed that the ANN modeling gives the best models for estimating LOS and capacity. Also, it is better for analysis to separate the desert and agricultural sites. In addition, the most influential variables on LOS and capacity in desert sites are HV and lane width (LW), respectively, while in agricultural sites are LW and existence of side access (SA), respectively. These results are so important for road authorities in Egypt as they can determine LOS and capacity for different tangent sections and improve the traffic performance of them in the future.

Keywords

• Multi-lane highways
• Heavy vehicles
• Level of service
• Capacity
• Artificial neural networks