Predicting the Severity of Accidents at Highway Railway Level Crossings of the Eastern Zone of Indian Railways using Logistic Regression and Artificial Neural Network Models

Road-railroad level crossing accidents pose serious safety risks to road users, and their significant increase requires more research efforts to propose substitute solutions. Such a solution must consider the impact of intersection geometry, user perception, traffic characteristics, driver behavior, environment, and seasonal variations on accidents. This study explores the considerable number of such accidents and develops a predictive model using all the factors that influence them. For these objectives, data were collected from databases maintained by the zonal head office of the East Central Railway (ECR) in India. Data included 175 level crossings that experienced at least one accident between 2006 and 2021 in the ECR region. This study presents two accident prediction models using logistic regression and ANN for the predominant factors of accidents in the ECR zone of Indian railways. The accuracy of fatal accident prediction was 96% for logistic regression and 98% for ANN.