Abstract:

In recent years, consumers ask for a higher demand on the security and quality of fruit and vegetable products. But the existing system can only predict the state of fruits and vegetables based on the temperature and humidity of the environment, and factors such as personnel operation and equipment are not taken into account. To solve these problems, we analyze the factors which affect the quality of fruits and vegetables in the cold-chain process, establish the early warning index system, and integrate the traceable and monitorable information in the supply chain. The BP neural network is utilized to build the security warning model, and we train it and make prediction. The results show that the proposed method has fewer errors than traditional time-series and regression analysis methods when solving practical problems, and can effectively increase the security risks and hazards warning accuracy of fruits and vegetables in cold-chain logistics.

Key words: cold-chain logistics；early warning；neural network