Lane-specific Traffic Flow Control

Traffic congestion on motorways has become a common phenomenon across the world. Congestion on motorways is a major discomfort to road users and causes many problems such as reducing the capacity of the motorways, increasing travel times and fuel consumption leading to high economic and societal costs. Alleviating congestion and providing better mobility is one of the major problems facing transport engineers around the world. Traffic management and control aims to solve, reduce or at least postpone the problem of congestion. On multi-lane motorways, disproportionate usage of one or more lanes can lead to congestion setting in on heavily used lanes while spare capacity is still available on other lanes. Disproportionate lane flow distribution is usually a result of the lane changing behaviour of drivers. After the onset of congestion, the roadway capacity reduces by approximately 5-30%, a phenomenon known as capacity drop. While traffic flow theory and control is a well-established scientific discipline, traffic control applications have mostly been limited to the roadway level regardless of how the traffic is divided over the lanes. Thus, there is a need for the development of effective traffic control measures on multi-lane motorways which require the treatment of different lanes of a motorway as independent but interacting entities. This dissertation aims at making the step to lane-specific traffic control to mitigate the negative effects of lane changing on the traffic flow in multi-lane motorway bottlenecks. The main objective of this dissertation is to develop lateral control measures which influence the lane changing behaviour of traffic on motorways and evaluate their impact on traffic flow efficiency. The central research problem is addressed from two directions based on the composition of traffic. The impact of lateral control on traffic flow efficiency in homogenous traffic (where all vehicles in the traffic are assumed to have similar characteristics and behaviour) is firstly investigated. With advancements in automation and communication technologies, intelligent vehicles are slowly making their way on the roads and interacting with conventional traffic. This calls for the development and testing of modelling and control approaches that are robust to the different types of intelligent vehicles and their corresponding market penetration rates. Therefore, the impact of lateral control measures on traffic flow for a mixed traffic stream of human driven and intelligent vehicles is evaluated. A macroscopic approach describing aggregate flows is adopted to deal with homogenous traffic while a microscopic approach targeting individual vehicles is used for mixed traffic to account for the variation in traffic composition...