Mostafa Reisi Gahrooei, Ph.D., Develops Data-Driven model to Detect Transportation Disruption During Extreme Events

UF Industrial & Systems Engineering Assistant Professor Mostafa Reisi Gahrooei, Ph.D., has received funding from the National Science Foundation for his research in developing a proactive, data-driven framework for monitoring road transportation networks during extreme events. The detection of abnormal traffic patterns based on incomplete traffic data and prediction of disruptions rely on advanced AI technologies, including tensor completion, deep neural networks, and self-expressive models.

Normal traffic patterns are altered by extreme events, such as natural disasters or large unusual man-made events.  These changes may result in disruptions that negatively influence emergency management processes such as evacuations, rescue and recovery operations. Prediction of these disruptions is critical for successful emergency management before, during, and after an extreme event.

“The benefits of a more efficient emergency management process include an enhanced quality of life, health and well-being for commuters and others that are using a specific infrastructure. My hope is that ultimately this research will lead to more sustainable and resilient cities that can function properly even under the stress of an extreme event,” said Dr. Reisi.

Unlike current methods for monitoring road networks, this new framework will not only monitor real-time traffic data for early detection of changes in traffic before, during and after an extreme event, but will also predict any disruptions in the network that this event may cause. These predictions would be at a road segment level of granularity, which means that while it will consider the entire network, it will predict the exact road that is going to be disrupted.

Dr. Reisi hopes this research will help transform current emergency management systems and will result in more efficient and successful rescue and recovery operations.