Dynamic traffic perception-based route optimization and adaptive control strategy for agricultural product cold chain distribution

Abstract

Urban-rural agricultural product cold chain distribution faces significant challenges from time-varying traffic conditions and the dual constraints of timeliness and temperature control. Traditional static routing approaches fail to adapt to dynamic traffic fluctuations, leading to delivery delays and increased cold chain disruption risks. To address these issues, this paper proposes a dynamic traffic perception-based route optimization and adaptive control strategy. A multisource traffic perception framework is constructed integrating floating vehicle data, roadside sensor information, and historical traffic patterns to enable real-time road condition estimation. Based on the perceived dynamic traffic states, a time dependent vehicle routing model with time windows and temperature constraints is developed. An adaptive genetic algorithm with elite immigration strategy is designed to solve the dynamic optimization problem, and a rolling horizon control mechanism is introduced to enable real-time route adjustments. Simulation experiments based on an urban-rural road network scenario demonstrate that compared with traditional static dispatching, the proposed strategy reduces average delivery time by 21.6%, improves on-time delivery rate by 15.3%, and decreases cold chain temperature excursion events by 61.8%. The results validate the effectiveness of dynamic traffic perception in enhancing cold chain distribution efficiency and reliability.