Firefighter fatalities and injuries in the U.S. remain too high and firefighting is too hazardous. Until now, firefighters do not have any tools except relying on their experience to avoid life-threatening fire events, such as flashover. The current state-of-the-art prediction models cannot be used in real-life firefighting because these models do not account for realistic conditions and require lengthy computational time.
In order to overcome the practical challenges and numerical bottlenecks, the Fire Research Division of NIST’s Engineering Laboratory is strived to develop next generation prediction approaches to achieve real-time, numerically efficient, accurate predictions in real fire situations. The candidate will develop novel machine learning frameworks to tackle various realistic challenges to achieve desired model performance to meeting real-life firefighting needs. Some of our latest works to overcome problems associated with lack of real fire data and flashover prediction forecast before the occurrence are included as reference [1, 2].
We are seeking scientists to work with us to develop machine learning based prediction frameworks to provide real-time, trustworthy, and actionable information to enhance situational awareness, operational effectiveness, and safety for firefighting.
 Tam, W.C., Fu, E.Y., Peacock, R., Reneke, P., Wang, J., Li, J. and Cleary, T., 2020. Generating synthetic sensor data to facilitate machine learning paradigm for prediction of building fire hazard. Fire technology, pp.1-22.
 Wang, J., Tam, W.C., Jia, Y., Peacock, R., Reneke, P., Fu, E.Y. and Cleary, T., 2021. P-Flash–A machine learning-based model for flashover prediction using recovered temperature data. Fire Safety Journal, 122, p.103341.
Deep Learning; Compartment Fires; Real-Time Forecast; Realistic Conditions; IoT Sensors