Investigating the Effect of Traffic Network Information on the Evolution of Traffic Flow in Intelligent Transportation Systems under Herd Effect

Intelligent transportation has emerged as a critical paradigm in the transportation sector, underscoring the growing significance of digital information. The extent to which travelers comprehend transportation network information fundamentally influences the dynamics of traffic flow evolution. Traditional random user equilibrium models assume that travelers possess knowledge of segment flow information; however, they fail to account for route flow information. To date, research has yet to investigate how travelers’ decision-making behaviors are altered following the acquisition of route flow information. When endowed with such information, travelers frequently demonstrate behaviors influenced by the bandwagon effect, adjusting their routes to conform to the choices of the majority. This behavioral modification disrupts the existing equilibrium, resulting in a continued evolution of traffic flow until a new stable state is achieved. To examine the implications of transportation network information on traffic flow evolution, this study proposes a generalized travel impedance function that incorporates the bandwagon effect, extending conventional route impedance frameworks. Subsequently, an entropy-based Logit-SUE flow evolution model is constructed using Kullback-Leibler divergence, with a comprehensive analysis of its properties and a demonstration of its consistency with the proposed generalized impedance function. Numerical analyses reveal that when travelers fully grasp route flow information, several salient characteristics emerge as traffic flow stabilizes: (1) The distribution of flow becomes concentrated on a limited number of paths, in contrast to the random user equilibrium; (2) The inertia exhibited by travelers has a negligible impact on the final evolution outcome; (3) The system’s total travel time is significantly lower than that of the random user equilibrium model. Furthermore, travelers equipped with transportation network information can optimize their route choices, enhancing flow distribution and minimizing overall travel costs. The findings of this research provide essential theoretical support and practical guidance for the advancement of traffic guidance software.