Machine learning-based incident duration prediction integrated with microscopic traffic simulation for expressway resilience evaluation.

This research framework tackles the tough problem of measuring how well expressways bounce back after incidents. It does this by combining machine‑learning predictions of incident duration with detailed traffic simulation. First, we collect past incident records and environmental data. Using that data, we train machine‑learning models—such as gradient boosting or random forests—to predict how long an incident will last. Next, we feed those predicted clearance times into a microscopic traffic simulator like SUMO or VISSIM. The simulator shows how congestion spreads over time and space, letting us see the impact on traffic flow. By running many incident scenarios and testing different response actions, the framework calculates expressway resilience with metrics

such as total delay, queue length, and recovery rate. The results give traffic‑management agencies clear guidance on how to improve incident‑response plans, design effective detours, and boost the overall reliability of urban expressway networks.

Message: Hi FE, I need help with “Machine learning-based incident duration prediction integrated with microscopic traffic simulation for expressway resilience evaluation.” in “Civil Engineering”