Real-Time Traffic Network State Estimation and Prediction withDecision Support Capabilities: Application to Integrated Corridor Management

This paper presents a real-time traffic network state estimation and prediction system with built-in decision support capabilities for traffic network management. The system seeks to provide traffic network managers with the capabilities to estimate the current network conditions, predict congestion dynamics, and generate efficient traffic management schemes for recurrent and non-recurrent congestion situations. The system adopts a closed-loop rolling horizon framework in which network state estimation and prediction modules are integrated. The system is applied in the context of Integrated Corridor Management (ICM), which is envisioned to provide a system-based approach for managing congested urban corridors. A genetic algorithm methodology is developed to generate efficient traffic management schemes that integrate preapproved control actions by all managing agencies. The system is applied to a section of a commuter corridor in Dallas, Texas. The results show the ability of the system to improve the overall network performance during hypothetical incident scenarios.