A Driving Safety Evaluation Algorithm Based on TOPSIS Model of Game Theory Combination

The road transportation industry is developing rapidly, but it is also facing a severe security situation. Drivers are the foundation and the source of traffic safety management, their bad driving habits have huge potential safety risks. In order to automatically assess the driving behavior safety, we propose a driving safety evaluation algorithm based on TOPSIS model of game theory combination. Firstly, according to the characteristics of driver behavior and the management requirements of the transportation industry, we extracted 6 first-level indicators and 12 second-level indicators about driving safety evaluation. Secondly, since the importance of different indicators for driving safety evaluation is different, in order to better reflect the importance of the above indicators, we use fuzzy analytic hierarchy process and entropy weight method to calculate the subjective weight and objective weight of the indicators respectively. Then, in order to scientifically optimize the combination of subjective and objective weights, so as to improving the rationality of indicator weighting, we use the game theory set model to optimize the combination of subjective and objective weights. Finally, we use an improved ranking method approximating the ideal point value (TOPSIS) to evaluate the driving safety to avoid the problem that those sample points which are equidistant from the ideal point and the negative ideal point cannot be sorted. The experimental results show that the proposed method can effectively evaluate the safety of the driver's driving behavior, and help transportation companies and industry management departments to timely detect and control the driver's irregular driving behavior.