Research And Application Of Driving Risk Assessment Method Based On Driving Behavior Data

In recent years,in order to adapt to the production methods and circulation needs of the modern economy,road transportation has developed rapidly,playing an increasingly important role in the national economy.However,while road transportation promotes social production,promotes economic development and improves people’s living standards,the road traffic safety problems it brings are becoming increasingly prominent.At present,road traffic accidents have become one of the most serious problems that threaten people’s public safety.The data shows that the number of cargo vehicles in the country has increased year by year while the number of passenger cars has rapidly declined.How to carry out effective road transportation safety management to reduce the occurrence of accidents is a major problem we are currently facing.Looking at the situation of road transportation in China,it is found that on the one hand,the road transportation is complicated,the transportation routes are long,and the scope is quite wide;on the other hand,the number of transportation operators(convoys)is large and scattered,and it is extremely difficult to manage safe transportation.Effective security monitoring and management is particularly important.As the most basic management unit of vehicle companies,there are many ways of vehicle fleet safety management,among which the most popular is driver behavior safety management."Human" is the main body to control the occurrence of traffic risks.According to the statistics of relevant data,more than 90%of traffic accidents are caused by improper driving behavior of drivers.Therefore,in the safety management of transport vehicles,the management of drivers’ driving behavior is particularly important.Therefore,this paper judges the driving risk of drivers through their driving behavior,promotes post-processing to prevention in advance,standardizes driving behavior according to driving safety standards and fleet management standards,reduces the probability of safety accidents of transport fleet,and achieves the purpose of safety management of transport fleet.The research content of this paper is:(1)The combination model is used to evaluate the risk of driver’s driving behavior.Through the Pingtong intelligent video monitoring and alarm system database,the daily driving behavior data of drivers in each operation team are obtained.After data cleaning and interpolation,the effective data are obtained.The driver’s behavior is included in the indicator system of driving risk assessment,which is divided into four categories:behavior habit category,"three urgent" risk category,vehicle alarm category and vehicle driving category.A data index system is established based on driving behavior data,and a combined model based on comprehensive evaluation method is proposed to estimate driving behavior risk.Entropy method is used for objective weighting and AHP method for subjective weighting.Entropy AHP method is used to get the optimal weight solution among four kinds of indexes.Fuzzy comprehensive evaluation method is used to get the risk score and ranking of driver’s driving behavior,which makes the combination model more scientific and accurate.Then the risk scores of the combination model are classified,and the main influencing factors are found out.(2)The driving risk score is applied to driver safety management of motorcade.This paper puts forward a fleet safety management method based on driving behavior safety score,applies the results to the actual cases,makes the combination of fleet safety management and driving behavior safety score,puts forward the management control evaluation method between them,and gives the corresponding relationship between driving behavior safety score and driver safety risk rating,so that the enterprise fleet can be graded by driving risk score And the main influencing factors to improve the driving standard.At the same time,quantifying the risk level of fleet in daily driving,making the hidden risks transparent and clear,effectively promoting the fairness,standardization and differentiation of enterprise fleet management,will play a role in promoting the reform of China’s transport industry market.The innovation of this paper is as follows:firstly,breaking through the limitations of existing research road,environment and vehicle factors,focusing on the bad driving behavior of drivers.The data studied in this paper is the driving behavior data of drivers,starting from the perspective of driving habits,combined with the driving habits of drivers,"three emergencies"(acceleration,braking and turning)dangerous,vehicle alarm type The second is to build a vehicle driving risk control model based on the survey data,and transform the analysis after the accident into the analysis before the accident,so as to reduce the probability of the risk and regulate the driver’s driving behavior.Finally,the model and its conclusions are applied to the fleet safety management to help the fleet enterprises to assess the driving risk in advance.The assessment results are of great significance to the safety management and risk avoidance of the fleet.There are still some problems and shortcomings in this paper:the installation of ADAS system is not universal,and the ranking of vehicles is likely to have some deviation;the ultimate purpose of studying vehicle risk is to calculate the possibility of its occurrence,and at present,the specific accident information of vehicles can not be obtained to give a basis for judgment.