The Research Of Logistics Demand Prediction System And Application Based On Gray Neural Network

Logistics demand forecasting is the basis of regional logistics system planning,effectively predict the trend of future material flow in the region is beneficial to maintain the balance between the supply and demand of local logistics resources.Therefore,the relevant factors of affecting logistics demand are in-depth analyzed and make the correct logistics demand forecast which can provide scientific data for regional logistics departments as a reference to grasp the intensity of regional logistics demand,which is also helpful to the regional government departments to formulate relevant development strategy for logistics industry.By analyzing the research achievements of current industry experts and scholars in the field of logistics demand forecast,to select effective prediction method of regional logistics demand for scientific evaluation,and to understand the development status of the logistics market,and take some effective measures to promote the healthy development of logistics industry in the future.Mainly from the following aspects of work:(1)Analyzes the factors that affect the scale of logistics demand,and optimizes the current index system of logistics demand forecasting.By using the grey correlation method to select the several effect factors of high correlation with the scale of logistics demand which based on the general index system and the actual prediction data of the local,and to construct the forecasting index system of local logistics demand based on these factors.(2)Considering nonlinearity,complexity and randomicity of the regional logistics system,the prediction of regional logistics demand is carried out by constructing the grey neural network model that combination of grey theory and BP neural network.The prediction method that combination of the two models not only the use of the unique advantages of the grey prediction method in dealing with incomplete information and small sample data,but also play the excellent learning ability of the BP neural network model in nonlinear system prediction.(3)Analyzes and designs the functional modules of the logistics demand forecasting system and completes the development of the system based on the grey neural network model and related theoretical knowledge.(4)The logistics demand forecasting system is used to forecast the logistics demand scale of Beijing,the results show that the system can effectively predict the change trend of logistics demand scale in Beijing in the short term and the results can be used as the basis of logistics planning and control in the future of Beijing.The logistics demand forecasting system has certain application value in real life.