"Motivated by the need to reduce congestion on the nation's roadway network, this research seeks to determine when a new detection scheme at actuated signalized intersections may help. This new detection scheme is called lane-by-lane detection. It "looks" at traffic in all lanes individually to determine the traffic volume on the roadway. Specifically, the problems addressed are investigating alternate signal timings and investigating the traffic volume's effect on performance of the detection scheme. For the alternate signal timings, an Evolutionary Algorithm (EA) was employed to identify the optimal signal timing. This algorithm creates signal timings and tests them in the microscopic traffic simulator, VISSIM. VISSIM returns the average delay to act as a measure of the signal timing. From this value, the EA uses evolutionary techniques found in biology to find the optimal signal timing. The investigation of traffic volume scenarios was performed using the optimal signal timings found by the EA as a starting point. This signal timing was run 25 times under each type of detection and the results were compared for 10 volume scenarios. A comparison of the performance for each volume scenario was completed. The results of the EA conclusively show that there is no signal timing which is better suited for lane-by-lane detection than for single-channel detection. Based upon this it is recommended that lane-by-lane detection be implemented with no other change to the signal control logic. Then the traffic signal volume study showed that low to moderate traffic volumes were necessary at the intersection for lane-by-lane detection to be beneficial. Also low left turning volumes were necessary on all approaches because large volumes sometimes interfered with thru traffic. If these guidelines are kept in mind, the reduction in delay per vehicle should be approximately 1.65 seconds"--Abstract, page iii.
Civil, Architectural and Environmental Engineering
M.S. in Civil Engineering
Missouri University of Science and Technology
viii, 61 pages
© 2009 Eric G. Strack, E.I.T, All rights reserved.
Thesis - Restricted Access
Library of Congress Subject Headings
Roads -- Interchanges and intersections
Traffic signs and signals
Electronic traffic controls
Traffic flow -- Computer simulation
Evolutionary programming (Computer science)
Print OCLC #
Electronic OCLC #
Link to Catalog RecordElectronic access to the full-text of this document is restricted to Missouri S&T users. Otherwise, request this publication directly from Missouri S&T Library or contact your local library.http://laurel.lso.missouri.edu:80/record=b10251215~S5
Strack, Eric G., "An evolutionary algorithm approach to actuated signal timing optimization under lane-by-lane detection condition" (2009). Masters Theses. 4512.