| With the popularization of Internet and information technology, the development of logistics should be more intelligent to create great value for the logistics industry. In the whole logistics process, an important node of logistics distribution center belongs to the logistics network, the upstream of the logistics distribution center is suppliers, and the downstream is distributors and consumers. In addition, the distribution costs occupy a considerable proportion of the total cost of logistics. How to choose the logistics distribution center effectively and scientifically is a problem worthy of study.As a tool to acquire, analyze, store and decide mass data rapidly and effectively, big data has some characteristics, such as the huge volume of data, fast speed in data storage and transmission, frequent interaction etc. According to many scholars, big data will realize the transformation from traditional logistics to modern logistics. This paper studies the optimizing logistics distribution center problem by using big data and MapReduce platform, combining with the data mining clustering algorithm. The specific contents are as follows:1. Big data and modern logistics. This part describes the concept, the characteristics of big data and modern logistics. It proposes a concept of "the big data of logistics", based on the contrast and analysis of the structure of big data and logistics data.2. The location of logistics distribution center based on big data. It explains the concept of logistics distribution center, and comparative the difference between modern logistics distribution center and traditional distribution center. Finally, it expounds algorithm of optimizing logistics distribution center in the age of big data.3. Design and improvement of DK-Means parallel algorithm based on the MapReduce. In reality, there is not straight line distance between nodes. However, Dijkstra algorithm can help. Therefore, we choose the DK-Means clustering algorithm. |
PDF Full Text Request