U.S. Dept. of Transportation: Hazmat Incident Prediction

Partner(s) U.S. Dept. of Transportation
DSWG Teammates Rocio Ng, Jude Calvillo, Wade Fuller, Tyler Field, Catherine Zhang, Daniel Schweigert, Ben Lucas

Project Detail

This was one of two tools the DSWG developed to answer the U.S. Department of Transportation challenge at Bayes Hack, 2016, which we ultimately won. It is a predictive model and product for predicting hazmat incidents and risk across/between the States, with interactivity for manipulating the model’s four most influential factors. It offers a pseudo-MSE of ~75%, in its alpha delivery, and we are currently working with the Department of Transportation in honing its data sources for greater granularity and precision. We are also refining its front-end for more intuitive temporal-spatial analysis.