A data-driven conceptual framework for understanding the nature of hazards in railway accidents

Abstract

Hazards threaten railway safety by their potential to trigger railway accidents. Whilst there are a considerable number of prior works investigating railway hazards, few offer a holistic view of hazards across jurisdictions and time and demonstrate policy implementation due to the inability to analyse a large amount of safety-related textual data. The conceptual framework HazardMap is developed to overcome this gap, employing open-sourced Natural Language Processing topic model BERTopic for the automated analysis of textual data from Rail Accident Investigation Branch (RAIB), Australian Transport Safety Bureau (ATSB), National Transportation Safety Board (NTSB) and Transportation Safety Board of Canada (TSB) railway accident reports. The topic modelling depicts the relationships between hazards, railway accidents and investigator recommendations and is further extended and integrated with the existing risk theory and epidemiological accident models. Results show that each hazard in the railway system has different aspects and could trigger a railway accident when combined with other hazards. Each aspect can be partially or fully addressed by implementing hazard mitigation policies such as introducing new technologies or regulations. A case study of the application to the risk at level crossings is provided to illustrate how HazardMap works with real-world data. This demonstrates a high degree of coverage within the existing risk management system, indicating the capability of helping policymaking for managing risks with adequate accuracy. The primary contributions of the framework proposed are to enable a huge amount of knowledge accumulated for an intuitive policymaking process to be summarised, and to allow other railway investigators to leverage lessons learnt across jurisdictions and time with limited human intervention. Future research could incorporate data from road, aviation or maritime accidents.