Quantitative Risk Assessment of Pedestrians

Train Accidents in the Colombo Division Railway Network

Abstract

Rail transportation stands as a critical mode for both passenger and freight movement in Sri Lanka, acting as a major catalyst for economic activity while linking different regions across the country. Train accidents cause tragic loss of life, injuries, and lead to noteworthy financial and social disturbances. Unfortunately, there has been a lack of comprehensive studies on railway accidents in Sri Lanka, likely due to restricted data accessibility. Therefore, advancing research in this area and gaining a deeper understanding of the underlying causes and patterns of train accidents is essential. The main objective of this study is to quantify pedestrian–train accident risk across the Colombo Division railway network and to determine the risk status of each line segment of the network. The Colombo division comprises 5 railway lines and includes 194 stations. Through segmentation of each railway line based on stations, a numerical assessment of risk is conducted for each segment. Statistical techniques, including bootstrapping and K-means clustering, are employed to classify the calculated risk scores into three categories: high, medium, and low. The reliability of these risk categories is validated using Kruskal-Wallis rank sum test and Dunn’s test with Bonferroni correction. The results indicate that, out of the total 194 segments, 165 are classified as low risk, 10 as medium risk, and 19 as high risk segments. Among these, the Coastal line has the highest count of high risk segments. Although, the Northern line and Kelani Valley line exhibit no high risk segments, Puttalam line has 1, while the Main line shows 7 high risk segments. These findings offer valuable insights to relevant authorities for identifying railway line segments that are highly vulnerable to accidents and for guiding government efforts in implementing safety measures for railway operations in Sri Lanka. Future research could enhance this study by including analyses of train accidents involving vehicles, animals, and other properties.