Effects of Autonomous Vehicles on Freeway Traffic Performance

Abstract

Autonomous vehicles (AVs) are smart transportation technologies that have drawn significant attention recently due to their rapid development and promising future. Dubai is trying to promote the use of AVs on its road network as it announced its future strategy to make 25% of its transportation automated by 2030. One of the major challenges that are expected to happen is the interaction between AVs and Regular vehicles (RVs) as the mode share, for AVs (percentage of AVs) would not be 100% in the early stages of adoption, and this interaction is not well-researched so far. The purpose of this study is to evaluate the impact of AVs on freeway traffic performance. The study considers a segment of E311 (Sheikh Mohamed Bin Zayed Road) freeway in Dubai as the test corridor for the study. A microsimulation software (VISSIM) is used to model and evaluate different scenarios. Different traffic demand to capacity ratios are evaluated by considering demand to capacity ratios. The results show that increasing AVs mode share increases the average speed and reduces average travel time and delay. Also, the impact of AVs on freeway performance is higher when the demand to capacity ratio is higher. The minimum effect is achieved when there is a 5% AVs and the demand to capacity ratio is 0.6 while the ultimate case is for 100% AVs and demand to capacity ratio of 1.2. In this case, the increase in speed is about 115%, the reduction in the average travel time is about 1.5%, and the average delay is lower by about 87%. The results obtained in this thesis represent a lower bound of what can actually be obtained, as the considered simulations assumed the lane width and capacity to remain the same. In real applications, more improvements can be achieved by designating some of the road lanes for AVs use only, at high mode shares of AVs. Such lanes have smaller width than regular lanes, which will increase the number of the lanes and road capacity.