Major Factor Affecting Freight Train Collision Severity At Road-Rail Level Crossing Using Multinomial Logit Model

Abstract

In this paper, significant factors of freight train collision severity at rail-road level crossing, have been identified using a multinomial logit model. For this purpose, both inventory and collision database of Iran Railway company for the period of 2009-2015 have been used. This research develops two multinomial logit models for injuries and fatalities by using SPSS statistical software. The results illustrate that “type of pavement” has the most impact among significant factors in fatality collisions severity model. Also, “type of car” has the most influence among significant factors in injury collisions severity model. As a conclusion, paying exclusive attention to resulted factors leads to save many lives. Besides, it is shown operational factors also reduce collision severity considerably.