Stop and go traffic is a specific type of congestion in which vehicles are forced to decelerate and travel at a low speed for some significant amount of time before accelerating back to a higher speed. Because these oscillations in the speed, density, and/or flow of the traffic have negative impacts on a highway's level of service, travel time, driver safety, and fuel economy, they are of particular interest to traffic researchers and engineers. Initially, microscopic factors such as lane-changing (Gazis1962, Munjal and Pipes 1971), asymmetric acceleration/deceleration of drivers (Newell 1965), and driver overreaction to external stimuli (Chandler1958, Herman 1959) were attributed as possible causes of stop-and-go waves. However, a discrepancy exists between the scale of microscopic parameters and the scale of oscillation characteristics, so macroscopic considerations have been pursued more recently. In this dissertation, we extend some of the existing research (Jin 2003, Zhang and Nie 2008, Zhang, Nie, Shen 2008, Zhang and Shen 2009) on the mechanism for generation and propagation of traffic waves through freeway networks.
We begin by modeling a portion of freeway with an on-ramp and subsequent merge lane drop as a simple network with merge and diverge junctions. The celebrated kinematic wave model proposed by Lighthill and Whitham (1955) and Richards (1956), as well as the Supply-Demand model of node junctions proposed by Daganzo (1994, 1995) and Lebacque (1996) allow us to consider the oscillation patterns that are produced as a result of the network topology and driver's route choice. Between two different levels of upstream demand, we see consistent regions in which solution types are identical; these findings are in line with previous research.
We then turn to micropscopic car-following models as a tool to further examine wave patterns on the on-ramp network. We employ two different models with three velocity functions, each of which admit a distinctive fundamental diagram. We also formulate a method of implementing a macroscopic merge priority scheme in a microscopic setting. The results suggest that microscopic factors may add to the oscillatory behavior that emerges; they also support the notion that choice of fundamental diagram is just as important as choice of model.
Next, we alter our network to emulate a ring road with merge and diverge properties similar to that of the on-ramp network. The closed system provides for additional system dynamics to occur as a result of interacting waves; the inherent variability of upstream demand over time better mimics real freeway traffic as well. We find four different solution types that arise and characterize the conditions under which each one will occur. We are also able to conjecture route choice ratios that lead to optimal dynamics for a network with given initial density.
Finally, we allow the route choice on the ring network to vary according to driver reactivity. Immediately upstream of a diverge, drivers are given an estimate as to the travel time on each possible route; this information then may impact their route choice. We see that, for various levels of reactivity and initial network density, traffic can attain a steady-state in terms of diversion ratio and/or traffic flow. Otherwise, the dynamics primarily show predictable wave patterns. Within this study we evaluate two route choice models and two methods for estimating travel time.
We finish by summarizing our work, relevant results, and potential directions for future research and application.
|Advisor:||Zhang, H. Michael|
|Commitee:||Guy, Robert, Temple, Blake|
|School:||University of California, Davis|
|School Location:||United States -- California|
|Source:||DAI-B 73/03, Dissertation Abstracts International|
|Subjects:||Applied Mathematics, Transportation planning|
|Keywords:||Freeway networks, Route choice, Stop-and-go traffic, Traffic oscillations, Vehicles|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be