Hopfield artificial neural network-based optimization method for selecting nodes of fresh agricultural products international logistics network

China is the biggest producer of agricultural products in the world. In China, the majority of agricultural products are produced by small-scale farmers currently, and the whole supply chain of agricultural products go through many links, including harvesting, grading, packaging, precooling, storage, transportation, distribution, retail, consumption and so on. This leads to some distinct features and potential problems in the agricultural products supply chain. Agricultural products supply chain has the remarkable characteristics of fresh agricultural products supply chains, that is, the seasonality, periodicity, regionality of production and the perishable nature of products make the supply chain very fragile, any potential and existing risks in any links will affect the effective operation of the whole supply chain, damage the performance of node enterprises and the whole supply chain. The Hopfield artificial neural network used in this paper conducts in-depth research and analysis on the site selection method of international coordination network nodes for fresh agricultural products. The Gray prediction model was used to predict the future demand trend of fresh agricultural products, and it was concluded that fresh agricultural products are the main demand for fresh agricultural products for consumers in the future. Taking the producer's original distributor location as the alternative site, we elaborated on the preparation and implementation process of the site selection of the cold chain coordination distribution centre and applied the improved CFLP model to solve the problem based on the fixed cost, transportation cost, spoilage cost, refrigeration cost, and other factors, and apply the Hopfield artificial neural network algorithm under the condition of satisfying the objective function and constraints, and finally get alternative for the optimal solution.