Abstract
Road traffic congestion on the cold chain logistics not only increase the cost and time, but also creates certain negative impact on the national carbon emissions. To fully utilize the traffic resources, this study has classified urban road traffic congestion and defined the various vehicle delivery speeds with dynamic congestion levels. Simultaneously, it has developed the cold chain products replenishment strategy by considering delivery route, multi-depot condition and even vehicle types, aiming to minimize the total cost and carbon emissions, and maximizing the cold chain products freshness. To achieve this, this study build up a multi-objective vehicle routing optimization model and designed a hybrid algorithm combining large-scale neighborhood search and NAGA-II. Through computational analysis, this algorithm effectively overcomes the weak local search capability of NAGA-II and efficiently solves multi-objective problems. Moreover, under the simulated random traffic congestion conditions, this model able to demonstrate relatively stable planning results and address complex road traffic situations. Finally, this study able to analyze the impacts of various replenishment strategies, by considering multiple depots and sensitivity coefficients of cold chain products from delivery objectives. The analysis results also provides valuable insights for actual cold chain logistics distribution industry.