Modeling Railway Crash Severity Using Multinomial Logit Method Based On Freight and Passenger Trains

Because of its convenience and low cost, rail transportation has become one of the most attractive modes of transportation for passengers and business owners alike. Rail transportation has been reduced in popularity due to accidents. This study investigates the factors that affect the severity of accidents for freight, passenger, and other rail vehicles in three levels of death, injury, and damage. A multinomial logit model (MNL) regression was used to model the Iranian railway accident data from 2006 to 2021. The model quantifies the effects of each variable on the severity of events by quantifying the results. The results of this study indicate that type of accidents (with pedestrians), type of accidents (with road vehicles), speed exceeding the allowable speed (equal to or exceeding 30 km/h), speed during the accident (equal to or exceeding 30 km/h), multilane rails and moderate variables are among the variables that have a greater impact on the severity of accidents than other variables, so investing in these areas improves rail safety. Because of budget constraints, this research work will be effective in attracting the attention of policy makers and transportation managers in this area.