Abstract:

High loss rate has always been a pain point in the circulation of fresh agri-products. As the “First Mile” of cold chains， pre-cooling is the key to affecting the quality of fresh agri-products. In recent years， the construction of pre-cooling facilities in producing areas has been promoted in China. However， due to the difficulty of constructing pre-cooling stations in traditional building forms in fields， centralized pre-cooling is mostly adopted in regions where conditions permit. However， centralized pre-cooling stations are generally far away from the producing area of agri-products. If their locations are not reasonable， the loss rate of fresh agri-products will remain high. Therefore， it is necessary to find the best location scheme of pre-cooling stations considering the “First Mile” loss of fresh agri-products， so as to effectively reduce the loss and logistics costs in the circulation of agri-products. Although little literature has considered the pre-cooling station location problem， only the number and location of pre-cooling stations are considered. However， the type and capacity of pre-cooling stations are not involved. Moreover， different pre-cooling methods for different types of fresh agri-products are not considered， and the “First Mile” loss from producing areas to pre-cooling stations is ignored. Economies of scale in the construction of pre-cooling stations are also not considered.In this paper， the loss of multi-type fresh agri-products after being picked in the logistics process of producing areas→pre-cooling facilities→logistics centers is considered， and corresponding pre-cooling facilities suitable for their pre-cooling needs is built. By considering the economies of scale in the construction and operations of various pre-cooling facilities， a mixed-integer linear programming model is developed to optimize the layout of pre-cooling facilities and determine the number， location， type， and capacity of pre-cooling facilities. A genetic algorithm is proposed to solve the model according to its characteristics. Finally， the effectiveness of the proposed model and algorithm are verified by numerical experiments and a case study on the optimal layout of pre-cooling facilities for fresh agri-products in Chengdu. The results show that the first-mile loss of fresh agri-products and the economies of scale in the construction of pre-cooling facilities will significantly affect the location of pre-cooling facilities. In practice， it is necessary to combine the characteristics of fresh agri-products and the technical and economic indicators of constructing pre-cooling facilities to optimize their locations.

Key words: fresh agri-products, pre-cooling facility, location, genetic algorithm