Abstract Manager

In an era of rising traffic volumes, urban expansion, and increased unpredictability due to climate change and societal demands, traditional reactive traffic management systems face increasing challenges in meeting current transportation demands. Congested corridors suffer from diminished travel reliability, increased collision risks, and significant economic losses. A predictive transportation management approach offers the ability to foresee traffic patterns and proactively implement interventions, preserving network capacity, improving safety, and delivering a better quality of life for travelers.

Building on the work presented at TAC 2021, the Ministry of Transportation of Ontario (MTO), in partnership with CIMA+ and Aimsun, has further advanced its evaluation of a potential predictive transportation management approach. Addressing the limitations of traditional reactive systems, the project explored proactive traffic management strategies enabled by real-time data integration and simulation-based forecasting. QEW Innovation Corridor from Toronto to Hamilton was chosen as the expanded pilot study area as it is complimentary to existing COMPASS infrastructure and other emerging new technologies.

Aimsun Live was used as the modeling tool to predict traffic conditions, assess potential issues, and evaluate intervention strategies in real-time. The overall process of developing the Aimsun Live model included: (1) calibration of the study area microsimulation model; (2) generating categorized travel patterns based on historical traffic data; (3) generating simulation-based traffic operations predictions in real-time (15-, 30-, and 60-minute every 5 minutes) based on the observed travel patterns; and (4) evaluating an appropriate response plan for positive mitigation.

The response plans, such as promoting local diversion routes and control plan changes were tested for multiple lane closure and typical congestion events. Overall, the response plans implemented showed a reduction in mainline travel time between 15-35 minutes during the peak closure period, compared to the do-nothing scenario. Furthermore, to mitigate the potential impact of diverted traffic on diversion routes, the dynamic control plans were implemented to minimize the local corridor delays that were utilized for diversion. The findings validated the scalability of predictive management, particularly its effectiveness in dispersing traffic along complex urban and inter-urban corridors.

Future phases aim to establish continuous 24/7 parallel operations, integrate further municipal and transit networks, and leverage mixed data sources for real-time optimization. This project represents a significant step toward a proactive, data-driven approach to transportation management, setting the stage for safer, more reliable, and adaptive mobility systems.