Vehicle routing Problem for cold chain logistics based on data fusion technology to predict travel time

Cold chain logistics requires low-temperature transportation, which consumes more energy and has higher distribution costs than ordinary logistics. Moreover, as the scale of cities continues to expand, traffic congestion is becoming more frequent. Therefore, it is particularly important to plan the distribution route reasonably. In this paper, we study the problem of cold chain logistics vehicle path planning based on travel time prediction. First of all, multiple connected routes with real-time changes in traffic conditions between customers in the road network were considered to describe the distribution scene. Second, a genetic algorithm-optimized backpropagation algorithm fused travel time predictions for road segments based on fixed detector technology and floating car technology to improve the accuracy of road segment travel time prediction. Then, based on the prediction of road segment travel time, a method for predicting the travel time of the route is proposed, and the actual road network is transformed into a travel time network for each customer. Finally, the vehicle routing problem in cold chain logistics was investigated using predicted travel time as input. This problem is formulated as a bi-objective model aimed at minimizing costs and carbon emissions. To address this problem, the Non-dominated Sorting Genetic Algorithm-II (NSGA-II) was proposed. The study provides support for cold chain logistics distribution companies to develop distribution strategies based on local environmental policies and their own operational conditions.