Study On Accident Causation And Safety Risk Analysis In Railway Transportation

There is huge number of elements in a complex system and the relationship between elements is very complex.A comprehensive and synthetical analysis of complex systems is of great significance to reveal the structural mechanism and evolutionary rule of the system,to optimize the structure and organization mechanism of the system,to control the development and changing trend of the system,and to ensure the normal and orderly operation of the system.The safety management of complex system is an important part of system management.It ensures the system to achieve the best safety level under the existing cost and technology constraints through risk identification before the event,risk control in the incident and restoration after the accident.Complex system safety management can effectively prevent abnormal operation and serious losses of the system after internal and external disturbances.Therefore,it has attracted the attention in recent years.The railway transportation system is a typical complex system.Railway transportation safety management requires effective control of a lot of workers,a large number of facilities and complex internal and external environment,which increases the complexity of the problem.Once the abnormal situation occurs in the railway transportation system,it could cause serious influence on the fast running train.It could disturb the normal transport organization or causes damage to the equipment in light degree,and could cause frequent safety accidents and even casualties in serious degree.This could cause serious damage to the railway transportation enterprises,and also seriously threaten the safety of passengers and goods.It is an important means to effectively improve the security of the system by analyzing and discovering the cause of the accident.To reveal the shortages and deficiencies of safety management according to historical accidents,and excavate the cause of accidents,can help railway transportation safety management decisions and ensure the safety operation of the system.This paper aims at quantitative analysis of accident causation and safety risk for railway transportation complex system combined with systematic analysis method.The purpose of this paper is to reveal the mechanism of the cause of railway transportation accidents and the rule of the evolution of the safety risk.The specific research contents are as follows.Firstly,a method for modeling and analyzing the cause network of a directed weighted railway accident is proposed,which is based on the complex network theory.And the 203 railway accident investigation reports published by British Railway Accident Investigation Branch(RAIB)from 2006 to 2014 as an example,we describes the extraction process of the chain of events from accident reports.As a result,an accident causation network which composed of 109 nodes and 260 directed weighted edges is established.Finally,through the analysis of the statistical characteristics of the network,the key factors,key causation links and the rules of accident causations are found.It is found that the network has the characteristics of scale-free networks and small world networks.The research results provide guidance strategy for accident prevention,and also reveal the difficulties of accident prevention.Secondly,The MC_Apriori algorithm for fast mining frequent itemsets is proposed.A standardized analysis method of association rules for railway accidents causal factors is designed.By the approach of system analysis,data analysis and preprocessing,203 accidents published by RAIB are analyzed.The data set of accident causal factors is obtained,and the association rules of the causation of railway accidents are found out.Finally,the network analysis and visualization of high-support association rules,high-confidence association rules,and all association rules are analyzed and visualized.The results show that the association rules of the accident causal factors show significant clustering characteristics.Thirdly,according to analyze the characteristics of railway transportation complex system,combined with dynamics of complex network,this paper defines the state of factors and the evolution process of factor states in railway transportation complex system,and further proposes a dynamic model of safety risk spreading.The simulation method of safety risk spreading on the directed weighted railway accident eastablished aforementioned is constructed.The safety risk spreading process under different parameters is simulated,the validity of the model is verified,and the dynamic spreading rule of safety risk is revealed.Fourthly,A Railway Accident Causal Factors Analysis and Classification System(RACFACS)is eastablished.The system definition based on RACFACS,the method of constructing the data set of accident causal factors,the method of determining Bayesian Networks structure,and the parameters learning of Bayesian Networks are proposed.Taking railway accident reports published by RAIB from 2014 to 2016 as an example,the process of constructing data sets of railway accident causal factors under RACFACS method is described,and the process of building the Bayesian Network structure and the process of generating node conditional probability table are proposed.The key factors and the sensitive factors of the accident were found out.By using the probability reasoning of safety risk,the method of identifying the subsequent possible risk events is proposed with simulating the abnormal situation occuring in the system.This paper proposes the modeling and analysis method of complex system combined the characteristics of railway transportation.The method can be applied to find the hidden troubles and identify the safety risk during railway transportation safety management,so then provide guidance and reference for the safe production of railway transportation.