Many fresh agricultural products (FAP) cannot achieve "smooth flow of goods" and "make the best value". FAP are perishable items and difficult to store and transport for a long time. Therefore, it is particularly necessary to study the structure of FAP supply chain. In this paper, firstly, it introduced the features of FAP supply chain and analyzed the existing types and structural characteristics of FAP supply chains. Secondly, taking the case of Shaanxi Province, this paper took advantage of principal component analysis (PCA) and cluster analysis to study the FAP supply chain of 10 cities. The results show that Xi'an belongs to consumer-oriented logistics city, it should mainly develop the logistics supply chain whose core is the third-party logistics or supermarket chains. At the same time, develop the logistics supply chain which is dominated by production and processing enterprises or wholesale markets as supplement. Tongchuan is an agricultural production-oriented logistics city, it should construct the supply chain that focus on production and processing enterprises and wholesale markets. Others are defined as logistics eclectic cities, they should focus on the development of agricultural production and processing enterprises according to different products in different cities. Finally, a environment configuration and platform of fresh agricultural products supply chain status monitoring was built based on the network of things to obtain the real-time information of products. The conclusions have certain significance to the selection of FAP supply chain in Shaanxi Province.

Abstract. Real-time status monitoring of supply chain logistics is the basis of intelligent supply chain implementation. Supply chain logistics is essentially a process of objects move in space over time. It could also be abstracted as a process of materials transferred through nodes. First, therefore, this paper establishes a element node model of supply chain logistics to analyze the information collection required by supply chain logistics. It then uses internet of things to study the method of environment configuration of supply chain logistics real-time status monitoring, and classify the information collection required in supply chain logistics into two categories. A data processing model of supply chain logistics status monitoring is also developed based on Auto-ID calculation, with regard to the second information cannot be obtained directly through the data collection. Finally, software platform is built to verify theories and methods drawn.