Journal of Shandong Jiaotong University

To address problems in the traditional precooling mode for fresh agricultural products, such as dispersed layout of precooling facilities, low resource utilization, and mismatched supply and demand, this study proposes a mobile precooling pool mode based on distributed storage. In this mode, standardized containerized precooling units serve as functional carriers to form a distributed precooling resource pool characterized by centralized and shared resources as well as flexible scheduling. Focusing on the location, number, and capacity configuration of mobile precooling pools, a coordinated optimization model for mobile precooling pool location and capacity is constructed with the objective of minimizing construction cost, operating cost, transportation cost, and loss cost of fresh agricultural products. A genetic algorithm is used for model solving. Taking the cold chain logistics network for fresh agricultural products in Yantai City as an example, the layout scheme of mobile precooling pools is obtained and sensitivity analysis is conducted on key model parameters. The results show that a total of 33 mobile precooling pools are planned in the study area. The overall layout aligns with the distribution of major production areas of fresh agricultural products and is adjacent to trunk traffic networks, ensuring good logistics accessibility. Among them, seven large mobile precooling pools are located in the core areas of each district or county, twelve medium mobile precooling pools are arranged at secondary traffic nodes, and fourteen small mobile precooling pools are located in relatively remote areas with concentrated production of specialty fresh agricultural products. When the loss intensity ratio of fresh agricultural products is smaller, losses during the ambient waiting-to-be-precooled stage at the origin are larger, and the model tends to deploy medium mobile precooling pools in a more dispersed manner to shorten the ambient retention time and reduce postharvest losses. When the gradient difference coefficient of fixed construction cost is smaller, the cost gap between small and medium precooling pools decreases while that between medium and large precooling pools increases, and the model prioritizes the deployment of medium mobile precooling pools. When the capacity decrement coefficient of unit variable construction cost is larger, the marginal cost of large mobile precooling decreases, and the model prioritizes the deployment of medium and large mobile precooling pools to achieve centralized and coordinated allocation of precooling resources. Through dynamic scheduling and hierarchical resource pool planning, the distributed mobile precooling mode provides an innovative path for intensive and refined layout of cold chain resources in the “first kilometer” of fresh agricultural products.