Study Of National Railway Loaded Wagon Flow Forecast System Based On Big Data Analysis Method

In the field of railway transportation, the wagon flow refers to a vehicle flow state in the railway network, and the vehicle is moving from the loading station to unloading station. Wagon flow forecast is to predict the process of this state. Accurate wagon flow forecast results are important prerequisite and the basic guarantee of railway transportation plan and adjustment. In the case of unified dispatching, researching the railway wagon flow forecast is significant for transportation dispatch and command. It has a certain practical significance to reduce the cost of transportation and improve the efficiency of the railway freight benefit. But China’s national railway network is huge and complex. The number of the railway station is large, the distribution is not uniform, the internal operation is complex, etc. and all these above bring a lot of difficulties to the accurate and efficient railway wagon flow forecast.With the deep development of informatization construction on China railway, a large number historical data about railway wagon flow has been accumulated. These data both in dimension and capacity is in accordance with the typical characteristics of big data. Therefore, it is proper that the big data analysis method is introduced into the railway wagon flow forecast, and it will become possible to forecast wagon flow efficiently and accurately.The thesis studies the goal, characteristics and effect of railway wagon flow forecasting. It finds the shortage of existing research, puts forward the method of big data of the railway historical data to study and analyze. We abstract the railway station by using the black box operation mode, use the computer simulation method to study and analyze the railway historical data by section, direction and the type of the operation, establish the simulation model and try to improve the accuracy and the operation speed.National railway network node is built up through the Hadoop system structure, the model and the algorithm help implement the system by using the object-oriented language to code. At last, the system achieve to rapidly process the wagon flow data and shorten the response time.