Developing Pedestrian Fatality Prediction Models Using Historical Crash Data: Application of Binary Logistic Regression and Boosted Tree Mechanism

Pedestrian fatality rate plays a key role in examining effectiveness of the road safety. The present study attempts to examine effect various categories accused vehicles and average 85th percentile speed at accident location on pedestrian crash fatality. also develop severity models using binary logistic regression boosted trees technique. Historical data, along with video recording technique sites, have been utilized for study. From equations, it is observed that when heavy vehicle (HV) hits as compared two-wheeler (2W), chance death increases 2.44 times. According Boosted tree model, contribution 60 %, whereas category 40 % prediction. should help planning better strategies like all red time intersections or foot over bridge critical locations.