Lane Changing Control to Reduce Traffic Load Effect on Long-Span Bridges

Abstract

Long span bridges are critical parts of a nation’s infrastructure network and congested traffic loading is the governing form of traffic loading. Groups of trucks travelling in conveys are created when faster moving vehicles, such as cars, change lane. In this research the authors investigate how the control of these lane-changing events can help reduce the traffic load effects on long span bridges. Real traffic data is used to simulate a traffic stream on a virtual road and bridge using a microsimulation model. Various lane-changing restrictions are examined and compared to the typical case of free lane changing. It is shown that restriction of lane changing is an effective means of reducing long-span bridge traffic load effect. This result may assist bridge owners in implementing measures to prolong the life of existing infrastructure. mance of the vehicle, and the psychological nature of the driver (including aggressiveness, perception, and reaction times). Yousif & Hunt (1995) describe some of these factors, giving an intuitive explanation of the pressure to change lane, in terms of the fundamental diagram of traffic (see for example, Gartner et al. 2005), shown here as Figure 1. Figure 1. Relationship between the fundamental diagram of traffic and the pressure to change lane (adapted from Yousif & Hunt 1995). Complicating the problem of understanding lane changing further, is the difficulty in obtaining experimental data. This is because the phenomenon is spatially and temporally distributed. Attempts to monitor individual vehicle trajectories have been made (see for example, Xuan 2006, who uses Global Positioning Satellites), but suffer from the fact that the driver is aware of the monitoring. The difficulty in determining individual vehicle’s lane-changing characteristics can be somewhat overcome using empirical observations. Yousif & Hunt (1995), Sparmann (1979), Brackstone et al. (1998), Hidas (2005) describe lane change rates as a function of flow, and these observations can be used to calibrate a lane-change model. Further efforts use other approaches, such as kinematic wave theory (Laval et al. 2006) and a gas-kinetic (Boltzmann) model (Helbing & Greiner (1997)). Hidas (2002 and 2005) outlines some recent work in the development of microscopic models for lane changing. The lane changing model used in this work is the microscopic MOBIL model proposed by Kesting et al. (2007) and explained later. 2 MICROSIMULATION MODEL 2.1 The Intelligent Driver Model The Intelligent Driver Model (IDM) is a microscopic car-following model developed by Treiber and others (Treiber et al. (2000a), Treiber et al. (2000b)). Its equations describe the motion of an individual vehicle in response to its surroundings, given some physically-meaningful mechanical and driver performance parameters. In particular the IDM is based on the idea that a driver tries to minimize braking decelerations. The acceleration a vehicle undergoes is defined by: