טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentMahmud Keblawi
SubjectSimulation Based Optimization of Adjacent Actuated Traffic
Signals
DepartmentDepartment of Civil and Environmental Engineering
Supervisor Professor Toledo Tomer
Full Thesis text - in Hebrew Full thesis text - Hebrew Version


Abstract

In traffic networks consisting of several signalized intersections, determining the traffic signal plan of each intersection appropriately while considering the other intersections on the network is a necessary condition for a reasonable level of service. As the transportation systems progressed, more sophisticated and sophisticated traffic signal plans have been developed that include many parameters, but to date no sufficient effort has been devoted to the development of evaluation and optimization methods that can cope with these complex plans .

This study presents the detailed structure of a simulation-based optimization system of complex actuated traffic signal plans for adjacent intersection. This system includes a mesoscopic traffic simulation model for modeling the movement of all road users at intersections and road segments that connect them, in order to examine different planning alternatives and the effect of the planning policy on the functioning of the intersections in the traffic network. In addition, this simulation model combined with a genetic algorithm (GA) will help determine the optimal parameter values of the signal plans for each intersection on the network .

By this integrated optimization system, three complex actuated traffic signals for three adjacent intersections in the city of Haifa was optimized. The results showed that the integrated optimization system has the potential to reduce delays in the transportation system, so that after the optimization process, the average person delay in these intersections reduced by 35% compared to the basic planning. The running times used in the optimization process showed that the mesoscopic simulation model has an advantage in terms of the running speed on the microscopic simulation models. In addition, comparing the optimization of single intersections to the optimization of adjacent intersections has found that considering the interdependence between the intersections in the traffic network may improve the quality of the solution.

The system makes it possible to adjust the optimal plan of each signalized intersection in the traffic network, accordingly, to the geometric and demand changes that occurs. In this work, the three intersection were optimized in 27 scenarios of changes in demand and their distribution, as well as changes in the lengths of the segments between the intersections. The optimization results of all scenarios showed that any change that occurs in the system directly affects the functioning of intersections across the network.