Industry Logistics Demand Forecast In Shanxi Province

In 2014,Shanxi Province gross industrial output value accounted for 42.9%.As the pillar industry of Shanxi Province,industry plays a very important role in economic development,thus demand projections for logistics industry is also the important industrial economic development and logistics management in Shanxi Province.Industrial logistics demand forecast results will affect the transportation network layout,industrial logistics infrastructure and the development of industrial enterprises in the future,so the Shanxi Province industrial logistics demand forecasting is the precondition for the healthy development of the logistics enterprises.Through the forecast of logistics industry in Shanxi Province,we can timely understand the demand of logistics industry in the province,thus it is reasonable to allocate resources to ensure that the logistics service supply and demand balance and improve the efficiency of economic activities.Industrial logistics demand forecasting,as the basis of the industrial economy and the activity runs,has always been the hot topic in scholars.They use different ways to construct model and the local logistics demand forecast,and a lot of articles use freight traffic indicates the amount of logistics.Preliminary studies with single prediction methods,including the moving average,growth rate,grey model and the stochastic time series,etc.Along with the social progress,economic activity is more complex and accurately forecast demand become more and more difficult,a single modeling method has been unable to meet the requirement for precision,so more research to the combination of a variety of methods will be the basis of the original model combined with some advanced models to forecast logistics demand.Machine learning model used by many scholars,Machine learning mode according to the dependent variable and influencing factors of relationship,through the combination of historical data and machine learning deduce the change trend of relatively more accurate,support vector machine(SVM)algorithm and BP neural network are two of the most commonly used.In the review of domestic and foreign research conclusion on the premise of the author for the issue,first of all,this paper summarizes the defects existing in the current study;Then the related concepts and the status of industry in Shanxi province are described.Secondly,according to the in Shanxi Province,the industrial product structure and industrial product transportation characteristics,and according to the strong correlation principle,obtain principle,comprehensive,practical principles from three aspects:the index of major industrial products,industrial products,transport index and industrial enterprise scale this three aspects selected 18 influence factors.According to using grey correlation method and principal component analysis to construct suitable for Shanxi Province industrial index system of logistics demand.Thirdly,introduce the least squares support vector mechanism deals and model;And optimizes the parameters of the kernel function through the particle swarm optimization(PSO);I use Matlab software to forecast shanxi industrial logistics demand;Finally,I summary and prospect in this paper.