Multi-objective collaborative optimization of urban traffic signal timing based on improved genetic algorithm and game theory model

Abstract

Driven by the swift pace of urbanization and the ongoing expansion of automobile ownership, congestion in city traffic has emerged as a major challenge hindering urban progress, causing heightened commuting delays, raised fuel usage, and worsened atmospheric pollution. Conventional fixed-time traffic signal management systems struggle to accommodate the fluid and intricate nature of urban vehicular flows, leading to suboptimal utilization of pavement assets. To address this issue, this study introduces a multi-objective cooperative optimization framework for urban traffic signal coordination utilizing an enhanced genetic algorithm (IGA) and game theory architecture. Initially, a multi-faceted optimization model for signal timing is constructed, prioritizing minimal average vehicle delay, maximal intersection throughput, and reduced carbon discharges as primary optimization goals. Then, the genetic algorithm is refined by incorporating adaptive crossover and mutation likelihoods along with an elite preservation mechanism to prevent early convergence and enhance global exploration capacity. Subsequently, the game theory framework is merged to harmonize the diverse interests of various intersections, converting the timing adjustment task into a non-cooperative game among nodes, and achieving collective regional traffic signal coordination via Nash equilibrium resolution. Ultimately, simulation trials are conducted utilizing actual traffic datasets from a prototypical urban roadway network through VISSIM simulation tools. Findings indicate that relative to traditional genetic algorithms and fixed-time control techniques, the suggested strategy decreases average vehicle delay by 28.3%, boosts intersection throughput by 19.7%, and lowers carbon emissions by 16.2%, which significantly enhances the functional efficiency of urban transportation systems and attains multi-objective collaborative optimization of traffic flow, energy efficiency, and ecological sustainability. This investigation offers a novel technical methodology for urban traffic signal regulation and advanced intelligent transportation governance.