J.W. Fang, P. Huang

Pages: 41-52

Abstract
It is urgent to study the comprehensive governance of traffic safety in a scientific and systematic way to further improve the governance level. To address the challenges posed by rampant illegal activities, inadequate road capacity, and high traffic accident rates in traditional comprehensive traffic safety management, the paper provides a novel method grounded in socio-economic and environmental composite factors. This approach establishes a comprehensive traffic safety prediction index system, integrating diverse socioeconomic and environmental elements, and couples these evaluation indices. By fusing the coupled indices with a grey-RBF neural network model, we forecast traffic safety values, which subsequently inform the formulation of tailored comprehensive traffic safety management strategies. Case analysis shows that after ten months of implementation, the number of illegal activities declined to 22, road capacity indices ranged from 0.79 to 0.88, and traffic accident rates fell within 2.8% to 6.8%. The results confirmed that the new method has promising practical effects.
Keywords: socio economic environment; composite factors; traffic safety; comprehensive governance; grey-RBF neural network model