This thesis establishes an intelligent cold supply chain management system which consists of two parts: one is the intelligent tracking system integrated with Radio Frequency Identification (RFID), Global Positioning System (GPS), and Wireless Sensor Network (WSN); the other is the cold supply chain model. This tracking system is mainly designed to monitor the food products during the transport, including two parts, a data terminal and a data server. The data terminal is installed inside a container, comprised of GPS, Bluetooth, industrial computer, WSN, RFID reader, RFID antenna, and Code Division Multiple Access (CDMA) modem. The data server is a computer which is able to access internet and has one Structured Query Language (SQL) database. Related application programs are developed with JAVA language. The whole system is successfully tested and meets the expectations we desired at the beginning. In this study, a refrigerator is used to simulate the environment of the container. The data terminal collects all information, including temperature inside the container, GPS location, Product's Identification, and current time in five minute intervals (customers will be asked to set this time interval at the beginning). CDMA cellular network provides the communication between the data server and the data terminal. The data server receives all information and saves the information in the SQL database, which can be used to predict the food safety. Advantages of this tracking system include the ability: 1) to trace and track the products starting from the suppliers to retailers; 2) to monitor and store important parameters during the processing and distribution of food products, such as temperature; 3) to communicate in real time for prompt response; and 4) to quantify food safety prediction. The objective of the model developed in this study is to maximize the profit of the cold supply chain. There are one distribution center, multiple retailers and suppliers involved in the cold supply chain. Since the real-time quality situations of products are available even during the transport, retailers can set prices of products based on the real quality situation. The company is able to dynamically plan the quantity of distribution from the distribution or suppliers' site. In addition, retailers are able to manage the inventory based on the real shelf life of products. This thesis also concludes all different inventory results for retailers under different scenarios which can help retailers to predict and manage the inventory. The optimization software, Lindo, is used to demonstrate that this model is capable to dynamically plan the distribution quantity. The sensitivity analysis for prices, transportation costs, and holding costs is discussed to simulate different situations during the transportation and distribution.