Doctor of Philosophy (PhD)
First Advisor's Name
First Advisor's Committee Title
Second Advisor's Name
Second Advisor's Committee Title
Third Advisor's Name
Third Advisor's Committee Title
Fourth Advisor's Name
Fourth Advisor's Committee Title
Fifth Advisor's Name
Fifth Advisor's Committee Title
Sixth Advisor's Name
Sixth Advisor's Committee Title
Non-recurrent Congestion, Traffic Signal Control, Data Analytics, High Resolution Data, Machine Learning, Optimization Method
Date of Defense
Improving arterial network performance has become a major challenge that is significantly influenced by signal timing control. In recent years, transportation agencies have begun focusing on Active Arterial Management Program (AAM) strategies to manage the performance of arterial streets under the flagship of Transportation Systems Management & Operations (TSM&O) initiatives. The activation of special traffic signal plans during non-recurrent events is an essential component of AAM and can provide significant benefits in managing congestion.
Events such as surges in demands or lane blockages can create queue spillbacks, even during off-peak periods resulting in delays and spillbacks to upstream intersections. To address this issue, some transportation agencies have started implementing processes to change the signal timing in real time based on traffic signal engineer/expert observations of incident and traffic conditions at the intersections upstream and downstream of congested locations. This dissertation develops methods to automate and enhance such decisions made at traffic management centers. First, a method is developed to learn from experts’ decisions by utilizing a combination of Recursive Partitioning and Regression Decision Tree (RPART) and Fuzzy Rule-Based System (FRBS) to deal with the vagueness and uncertainty of human decisions. This study demonstrates the effectiveness of this method in selecting plans to reduce congestion during non-recurrent events. However, the method can only recommend the changes in green time to the movement affected by the incident and does not give an optimized solution that considers all movements. Thus, there was a need to extend the method to decide how the reduction of green times should be distributed to other movements at the intersection.
Considering the above, this dissertation further develops a method to derive optimized signal timing plans during non-recurrent congestion that considers the operations of the critical direction impacted by the incident, the overall corridor, as well as the critical intersection movement performance. The prerequisite of optimizing the signal plans is the accurate measurements of traffic flow conditions and turning movement counts. It is also important to calibrate any utilized simulation and optimization models to replicate the field traffic states according to field traffic conditions and local driver behaviors.
This study evaluates the identified special signal-timing plan based on both the optimization and the DT and FRBS approaches. Although the DT and FRBS model outputs are able to reduce the existing queue and improve all other performance measures, the evaluation results show that the special signal timing plan obtained from the optimization method produced better performance compared to the DT and FRBS approaches for all of the evaluated non-recurrent conditions. However, there are opportunities to combine both approaches for the best selection of signal plans.
Previously Published In
Tariq, M. T., Massahi, A., Saha, R., & Hadi, M. (2020). Combining Machine Learning and Fuzzy Rule-Based System in Automating Signal Timing Experts’ Decisions during Non-Recurrent Congestion. Transportation Research Record, 0361198120918248.
Tariq, M. T., Hadi, M., & Saha, R. (2021). Using High-Resolution Controller Data in the Calibration of Traffic Simulation Models. Accepted for the 100th Annual Meeting of the Transportation Research Board, Washington DC, USA.
Tariq, Mosammat Tahnin, "Data Analytic Approach to Support the Activation of Special Signal Timing Plans in Response to Congestion" (2020). FIU Electronic Theses and Dissertations. 4546.
In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/
This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).