Research On Logistics Demand Forecasting Of Hebei Province Based On Lasso-BP Neural Network

In the current trend of economic globalization and regional economic integration,as the "accelerator" of economic growth,logistics has become increasingly important.Facing the increasing uncertainties in the world economy and the increasingly severe competition situation,strengthening regional cooperation has become a common choice for policy makers.Logistics can promote regional interconnection and free flow of resources,play a role in regulating supply and demand and enhancing regional industrial competitiveness.To maximize the role of logistics in promoting economic development,we need to build a modern logistics service system,and the forecast of logistics demand is an important basis.Through scientific forecast of logistics demand,adjust the supply-demand relationship timely according to actual demand,plan the logistics network reasonably,improve and optimize the logistics service system continuously and maximize the role of logistics in promoting social and economic development.At present,there are still some problems in the development of logistics industry in Hebei Province: unbalanced supply and demand of logistics market,extensive development model,low degree of industrial informatization,low efficiency and high cost of logistics.These problems seriously restrict the development of logistics industry in Hebei Province to a high-quality stage.Therefore,this paper establishes Lasso-BP neural network model to forecast logistics demand in Hebei Province,which provides a basis for integrating logistics resources and guiding the development of logistics industry in Hebei Province scientifically and reasonably.Based on the research of logistics demand forecasting at home and abroad,combined with the current situation of logistics development in Hebei Province,this paper studies the logistics demand forecasting in Hebei Province.Firstly,this paper analyses the influencing factors of logistics demand from four aspects of regional economy,regional industry,regional environment and other influencing factors,and uses correlation analysis method to analyze the relationship between various factors and dependent variables.On the basis of previous studies,this paper summarizes and constructs a more reasonable and perfect index system of logistics demand prediction in Hebei Province.Secondly,taking freight volume as the quantitative index of logistics demand,Lasso-BP neural network model is established topredict freight volume in Hebei Province.First,Lasso method is used to select four independent variables which have the most significant influence on dependent variables:added value of the second industry,investment in fixed assets of the whole society,railway operating mileage and highway mileage,and then the four independent variables are used as input variables to predict the freight volume in Hebei Province by using BP neural network.The results show that Lasso-BP neural network model is more efficient than the second exponential smoothing prediction model and the single BP neural network model.Thirdly,we use the Lasso-BP neural network model to forecast the freight volume of Hebei Province in the next five years.The results show that the freight volume of Hebei Province will increase steadily in the next five years,but the growth is not obvious compared with previous years,reflecting the problem of insufficient power of logistics demand growth.Fourthly,this paper puts forward the following suggestions for the development of logistics in Hebei Province from different perspectives:(1)guarantee the amount of investment in fixed assets of the whole society,especially in the construction of logistics infrastructure.Through increasing financial input,we can continuously update and improve the logistics infrastructure,improve the level of logistics hardware,boost the development of logistics industry from equipment and facilities,enhance the consistency between logistics industry and regional economic development,and enhance the growth power of logistics demand from social investment.(2)optimize the economic and industrial structure.Maintain the stable development of the secondary industry,increase the support to the real economy,and stabilize the supporting role of the secondary industry to the growth of logistics demand.At the same time,we will continue to support the development of the tertiary industry,release industrial vitality and increase new momentum for the growth of logistics demand.From the stable support and the increase of new momentum,we can promote the continuous growth of logistics demand.(3)strengthen logistics supply capacity.Perfect logistics service system,improve logistics service level,innovate logistics service products,provide more diversified and professional logistics services according to market changes.While meeting the existing market logistics demand,new logistics demand is generated through supply innovation,and the growth power of logistics demand is enhanced from the supply side.(4)emphasis should be placed on the cultivation of logistics talents.The development of all industries needs the support of talents.Universities and logistics enterprises should strengthen cooperation,give full play to their respective advantages,establish a joint training mechanism for talents,pay attention to training professionals with excellent professional quality and strong practical ability,and promote the sustainable and long-term development of logistics industry with talents.The innovation of this paper lies in the combination of Lasso's variable selection and BP neural network's non-linear prediction in the field of logistics demand forecasting,and the empirical analysis of logistics demand forecasting in Hebei Province is realized.By establishing Lasso-BP neural network model,it provides an effective way for government and enterprise to forecast logistics demand.