Research on Multi-Dimensional Pricing Model of UBI Auto Insurance Based on Fuzzy Controller

Aim: ing at the problem of usage-based insurance (UBI) auto insurance pricing, this paper proposes a multi-dimensional pricing model based on fuzzy controllers. In order to comprehensively consider the influencing factors of the rate, an index system of autonomous underwriting coefficient is proposed from four dimensions of driving distance, driving time, driving speed and driving behavior. A sub-fuzzy controller based on dimension index is proposed. The adaptive weight and learning factor are introduced into simulated annealing particle swarm optimization (SAPSO) algorithm, and combined with Metropolis criterion, the adaptive SAPSO algorithm is improved. In order to reduce the subjective defects of fuzzy controller, standard particle swarm optimization (PSO), parameter adaptive PSO and SAPSO algorithms are used for optimization respectively. The fitness function is improved to incorporate stability into the screening criteria. After comparison, it is found that SAPSO algorithm can get the optimal results in fuzzy controllers with lower dimensions, and parameter adaptive PSO algorithm in fuzzy controllers with higher dimensions can get better results. The optimal output result is selected and input into the total fuzzy controller to calculate the autonomous underwriting coefficient. The experimental results show that the multi-dimensional auto insurance pricing model based on fuzzy controller in this paper can better measure the driving risk comprehensively. It has higher accuracy and stability in rate determination.