Cold chain logistics(CCL) plays an increasingly important role in modern logistics. In order to achieve efficient and high-quality CCL, providing feasible methods in vehicle transportation and distribution(VCD) is crucial. This article adopts an improved ant colony algorithm(ACA), with transportation cost and time as the main objectives, to optimize common path selection problems in VCD processes, and proves the effectiveness of this algorithm through experiments. The improved ACA is introduced to enhance the search ability, and the pheromone is constantly updated in the search process to find a better path. The experimental results show that the total cost of CCL transportation for agricultural products obtained by improving the ACA is 821 yuan, and customer satisfaction reaches 100%. The advantage of this algorithm is that it can comprehensively consider various factors, search for global optimal solutions, adapt to various complex scenarios, and is not easily trapped in local optima.