A network accident causation model for monitoring railway safety

Abstract In railway systems, risk monitoring and accident causation analysis are important processes towards operational safety. This paper divides accident causal factors in a railway system into several error types, such as human and signal, and proposes a model based on a complex network for risk monitoring, where the risks of accident causal factors are quantified. This network accident causation model is used to identify accident causal factors and analyze how these factors affect each other, for example, how a signal error leads to a collision between two trains. The results of this case study show that in a complex environment, the proposed model can better identify the root causal factors by quantifying the accident causal factor risk, to find the causation chain based on the interactions among accident causal factors. Based on the analysis results, we can timely and correctly monitor the accident causal factors which have high possibility to raise faults or accidents, thereby protecting the railway system from these factors. The proposed network model provides an effective support for risk monitoring in a railway system.

[1]  Neville A Stanton,et al.  Exploring the psychological factors involved in the Ladbroke Grove rail accident. , 2011, Accident; analysis and prevention.

[2]  Natasa Przulj,et al.  Network analytics in the age of big data , 2016, Science.

[3]  Carl Macrae,et al.  Early warnings, weak signals and learning from healthcare disasters , 2014, BMJ quality & safety.

[4]  Wan Chul Yoon,et al.  An accident causation model for the railway industry: Application of the model to 80 rail accident investigation reports from the UK , 2013 .

[5]  Carole Duval,et al.  Methodological developments for probabilistic risk analyses of socio-technical systems , 2009 .

[6]  Philippe Lacomme,et al.  Competitive Memetic Algorithms for Arc Routing Problems , 2004, Ann. Oper. Res..

[7]  Eckehard Schnieder,et al.  Verification of the safety communication protocol in train control system using colored Petri net , 2012, Reliab. Eng. Syst. Saf..

[8]  M E J Newman,et al.  Finding and evaluating community structure in networks. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[9]  Andrew S McIntosh,et al.  Understanding the human factors contribution to railway accidents and incidents in Australia. , 2008, Accident; analysis and prevention.

[10]  Lars Harms-Ringdahl,et al.  Relationships between accident investigations, risk analysis, and safety management. , 2004, Journal of hazardous materials.

[11]  Doohee Nam,et al.  Accident prediction model for railway-highway interfaces. , 2006, Accident; analysis and prevention.

[12]  Keping Li,et al.  Railway accidents analysis based on the improved algorithm of the maximal information coefficient , 2016, Intell. Data Anal..

[13]  Ziyou Gao,et al.  Fault tree analysis combined with quantitative analysis for high-speed railway accidents , 2015 .

[14]  Liu Hong,et al.  STAMP-based analysis on the railway accident and accident spreading: Taking the China-Jiaoji railway accident for example , 2010 .

[15]  Lina Bertling Tjernberg,et al.  An Artificial Neural Network Approach for Early Fault Detection of Gearbox Bearings , 2015, IEEE Transactions on Smart Grid.

[16]  Neville A Stanton,et al.  Modelling of human alarm handling response times: a case study of the Ladbroke Grove rail accident in the UK , 2008, Ergonomics.