Masters Theses

Abstract

"The variable speed limit (VSL) system has been adopted on more and more roadways across the globe as an engineering tool for controlling traffic flow, particularly during congestion situations. An important and fundamental question raised among traffic engineers is: Are the costs of installing and operating variable speed limits really justified? In order to answer this question, this thesis measures the effectiveness of the VSL system applying the VISSIM microscopic simulation model to a non-recurrent congestion situation. For the analysis, a section of interstate was modeled in VISSIM and calibrated so that the simulated results closely reproduce the observed data collected from traffic detectors. The calibration process is accomplished using a genetic algorithm (GA) programmed in MATLAB. Once the interstate system is calibrated to the static speed limit condition, the VSL system is portrayed on the model by temporarily blocking lanes to measure the effectiveness of VSL during an incident situation. Multiple incident locations and different VSL configurations are tested to investigate various possible forms of reality. The simulated results of with-VSL and without-VSL systems were compared and evaluated.

The results show that no significant changes in travel times are observed, but queue lengths and fuel consumption decrease by 10-20% and 3-5% respectively. The speed variances upstream of the incidents also decreased 10-25% depending on location, which in the end may lead to fewer crashes.

Before a VSL system is constructed a study should be conducted to determine if the cost of installing and maintaining the system is justified by the benefits. The benefits of a VSL system include a decrease in fuel consumption as well as a decrease in the deviation of speeds, potentially leading to a reduction in crashes. Further studies need to be done to determine exactly how much the number of crashes could potentially be reduced. If a governing authority has the means for a VSL system they could potentially be justified by a reduction in crashes on the roadway, but the breadth of this thesis does not have enough evidence to support the case that there will be a definite reduction in the number of crashes"--Abstract, page iii.

Advisor(s)

Baik, Hojong

Committee Member(s)

Park, Byungkyu
Burken, Joel G. (Joel Gerard)

Department(s)

Civil, Architectural and Environmental Engineering

Degree Name

M.S. in Civil Engineering

Sponsor(s)

Missouri. Department of Transportation
Crawford Bunte Brammeier Traffic and Transportation Engineers

Publisher

Missouri University of Science and Technology

Publication Date

2010

Pagination

xii, 91 pages

Note about bibliography

Includes bibliographical references (page 90).

Rights

© 2010 Brian Joseph Schaefer, All rights reserved.

Document Type

Thesis - Open Access

File Type

text

Language

English

Subject Headings

Traffic flow -- Computer simulationExpress highwaysRoads -- Interchanges and intersectionsEvolutionary programming (Computer science)

Thesis Number

T 10280

Print OCLC #

870998458

Electronic OCLC #

909611523

Download

Share

 
COinS