Research On Key Technologies For Cause Analysis And Early Warning Of Road Traffic Accidents

With the rapid development of social economy,improvement of road traffic construction and increase in the number of means of transportation,the traffic accidents are becoming more and more serious,which cause great harm and loss to personel safety and property.How to reasonably analyze traffic accidents,explore the internal causes of accidents and reduce accidents,timely warning of accidents and prevent accidents before it happens has become an urgent problem in the field of transportation.This thesis takes road traffic accidents in Karamay as the research object,firstly,elaborates the ecological construction of big data of road traffic accidents,and then promotes the value of data through data sharing.Additionally,the general characteristics of accident distribution in the city are analyzed from the aspects of people,vehicles,roads and environment.A time-space analysis model of traffic accidents is established from the perspective of time and space,the road accident-prone time period and accident-prone road section are obtained.Using data mining methods to analyze the causes of accidents and early warning of accidents is the main content of this thesis.The occurrence of road traffic accidents has many factors,randomness,and contingency.This thesis excavates and analyses the causes and severity of accidents.The severity of accidents is classified into property loss accidents,injury accidents and death accidents.By using association rule,decision tree,clustering,and other methods to analyze the influencing factors of road traffic accidents from multiple perspectives and levels,a road traffic accident analysis system based on data mining is constructed.Furthermore,the relationships between traffic accidents and multi-dimensional accident factors are explored.Finally,the early warning indicators system of road traffic accident and the early warning model of accident are established by combining the analytichierarchy process with the fuzzy comprehensive evaluation method to realize the identification of road safety level.This thesis studies the technology of big data and data mining to provide effective assistant in decision-making for road traffic safety management.It also provides effective early warning for roads with low safety level.To a certain extent,it gives warning to traffic management departments,so that they can take timely measures to prevent accidents.This research has achieved some results in the practical application of road traffic accident management in Karamay,which has reduced traffic accidents and the losses caused by accidents in this city.