The Research And Design Of Business Intelligence System Based On Road Transport

With the constant improvement level of the enterprise informatization, the industry competition is getting more and more severe, especially in the background of the economic crisis. The rapid development of the high-speed railway brings about unprecedented pressure to the road transport industry. The road transportation enterprises in China shoulder the social responsibility and must pay attention to their own economic benefits simultaneously. Therefore, constructing a suitable business intelligence system for the characteristics of public transportation enterprises is of great significance. The building process of the the road transport business intelligence system is described in detail with the combination of the characteristics of the road transport, wihich is based on the road transport information system. Then the research on the optimization of the data warehouse system is conducted in this paper.Firstly, the project status, system architecture and the limitations of the road transport enterprise informatization system are analyzed. The building process of the requirement analysis, logic model and physical model is given in detail based on the platform of the sql 2005 business intelligence toolkit with the combination of the characteristics of business intelligence technology. Then, the data is extracted, converted, integrated and loaded in the warehouse. The multidimensional data set is created by using the SSAS, and the cluster analysis and prediction of the passenger volume by using K-means algorithm and the Markov prediction method.Secondly, Aiming at the problems of large volume of data and the complexity of the query condition, the S-tree, which is used in the traditional layered bitmap index structure, is abandoned. Then the clue red back tree with weight is introduced and the improved layered bitmap index structure is proposed. Moreover, the corresponding algorithm is given. At the same time, the efficiency analysis is conducted to evaluate the high effectiveness of the improved algorithm in processing the query.Finally, the ticketing situation is analyzed in detail by OLAP, and the effectiveness of the K-means algorithm and the Markov prediction method in the the classification and prediction of the passenger volume is verified.