Study On Bohai Economic Zone Logistics Demand Forecasting Based On Partial Least Squares-neural Network

With the rapid growth of economy and the accelerating informationization,as a new industrylogistics has rapidly developed.Logistics demand exists in all walks of life since the day oflogistics appeared.And because of the development of economy, the third party logisticscompanies are increasing.They outsource the logistics service of many enterprises.So logisticsdemand is growing. This rapid growth will likely bring problems:the logistics actual supplycapacity can not meet the logistics demand.The imbalance have the opposite effect todevelopment of economic and logistics.Therefore, forecasting the logistics demand isparticularly important.As the economy third "growth pole" of China,in support of national policy,the economy of Bohai economic zone develops rapidly.The overall strength significantlyincreases, opening to the outside world further expands,and the third industry developmentaccelerates. This paper will carry on the forecast to the logistics demand of Bohai economiczone.The prediction results can provide necessary basic data and decision support for thedevelopment of logistics in the region.In this paper, the establishment of partial least squares-neural network combination modelforecasts the logistics demand of Bohai economic zone.The first chapter, the introduction part,introduces the research background and significance of regional logistics demand forecasting,analyzes the domestic and international research status,and describes the main research contentsand the basic structure of this paper, technical route and innovations. The second chapterintroduces the concept and characteristics of regional logistics and regional logistics demand,analyzes influence factor of the regional logistics demand, and establishes the index system ofregional logistics demand forecasting. The third chapter discusses the emergence, developmentand application of partial least squares regression, expatiates the basic principle and algorithmderivation,and briefly introduces the auxiliary analysis technology of the partial least squaresregression.The fourth chapter introduces the BP neural network prediction method, expounds theconnotation,application and development of the neural network,analyzes the feasibility of theneural network being applied in regional logistics demand forecasting, and establishes regionallogistics demand forecasting model based on BP network. Finally combining the partial leastsquares regression method with BP neural network, establishes the combination model. The fifthchapter first briefly introduces the present condition of the development of economy andlogistics of Bohai economic zone, then the simulates and predicts Bohai economic zone logisticsdemand.The sixth chapter puts forward reasonable countermeasures of development of Bohaieconomic zone logistics according to the prediction results from the government level andenterprise level. The seventh chapter is the summary of this research work and the prospect of the further research.In the planning process of logistics system logistics demand forecasting is a very importantlink.In the logistics demand forecast, using the combination forecasting method of partial leastsquares-neural network can relatively scientifically forecasts the logistics demand trend in thefuture to use in practical.