Research On Railway Traffic Accident Prediction Based On HFACS Model And Grey System

In recent years,with the improvement of railway construction and operation level in our country,the layout of road network changes with each passing day.Especially,the application of a lot of technical equipment has improved the safety and reliability of transport.But with the high integration degree and automation of equipment,the difficulty of personnel operation is further increased.According to statistics,the high number of accidents is mainly related to human factors.At the present stage,Chinese academic circles have not discussed this topic as the core,and relevant conclusions are scarce.The human factor analysis model with multi-level perspective has not been formed,and the multi-factor accident prediction system has not been established.Therefore,it is of great significance to establish the human factor analysis model of railway traffic accidents,explore the human factor mechanism of railway traffic accidents and further predict the accident trend.Based on the theory of HFACS model and the current situation of railway transportation safety,this paper defines the specific content of human factors in railway transportation,studies the relationship between human factors and the number of accidents,predicts and validates the number of railway traffic accidents with the prediction model,and finally puts forward targeted improvement measures.Firstly,by analyzing the main causes of traffic accidents and collecting and comparing the research trends of traffic accident prediction methods,it is determined that the railway traffic accident prediction needs to combine HFACS model and grey system.Secondly,according to the classical HFACS model,according to the current calculation basis and logic of railway transport accidents,we comb the human factors.Then combined with the popular HFACS theory in present academia,we optimize the model structure again,in order to make it more valuable and significant in this paper.On this basis,it can collect railway traffic accident data more comprehensively and accurately,and carry out in-depth discussion on the statistical data by combining Chisquare test and concession ratio analysis,and finally get 13 causal chains of railway traffic accidents.Thirdly,according to the research method of grey correlation method,the correlation values of human factors of 17 railway traffic accidents obtained above are calculated and sorted,so as to obtain the influence degree of each human factors on the occurrence of railway traffic accidents.Finally,the human factors with high correlation degree were simulated and analyzed by combining grey prediction models GM(1,1),GM(1,N)and multiple linear regression models,and the prediction model with the highest accuracy was obtained.Then,the prediction model is used to predict,and finally combined with the result of the rise of the forecast data,the solution measures are developed for the human factors with high correlation degree.