Henan Province Logistics Demand Forecast Based On Improved Gray Neural Network

The rapid and diversified development of the social economy has broadened the development space of the logistics industry and promoted the diversification and complexity of the logistics industry.Since ancient times,Henan Province has always been a transportation hub of China.It has rich resources and is a very important city in central China.It is a key place for China to carry out logistics layout planning.Henan Province has also vigorously developed the logistics industry in recent years.Under this background,the decision-making planning of the logistics industry is particularly important for the development of the logistics industry.The forecast of logistics demand is the process of analyzing,inferring and speculating the future state of logistics demand.Accurate forecasting of logistics needs can provide a scientific basis for the decision-making of the logistics industry and play an important role in the development of the logistics industry.Therefore,it is imperative to study the logistics demand forecast of Henan Province.Taking Henan Province as an example,this paper conducts an empirical study on the forecasting of logistics demand in Henan Province.Firstly,it summarizes the domestic and international research on logistics demand forecasting,and then summarizes the logistics demand and logistics demand forecasting methods.It analyzes five aspects:regional GDP,industrial structure,household consumption level,domestic and overseas trade and logistics infrastructure construction.The main influencing factors of logistics demand were selected from Henan Province's GDP,primary industry output value,secondary industry output value,tertiary industry output value,household consumption level,regional consumer goods retail sales,total import and export volume,and fixed asset investment.Eight indicators are used as indicators of the influencing factors of logistics demand forecasting.The correlation degree analysis method is used to select the freight turnover volume as the quantitative index of logistics demand,and the index system of logistics demand forecasting is constructed.The characteristics of non-stationary time data sequence analysis can be analyzed by using ARIMA model.The gray GM(1,1)model is improved,so that the gray GM(1,1)model can study the data samples with complex rules;and the excellent nonlinear approximation ability of the BP neural network is used to improve the gray GM(1,1)the shortcomings of the model in approximating complex nonlinear functions,creating improvements Grey neural network(GM-ARIMA-BPNN)combined forecasting model;three models are used to simulate and predict the logistics demand of Henan Province,and then the simulation prediction results are compared to prove the prediction accuracy of GM-ARIMA-BPNN combined forecasting model.The highest,the smallest error.Then use GM-ARIMA-BPNN model to predict logistics demand in Henan Province from 2018 to 2020.Finally,according to the forecast results,this paper puts forward four suggestions on establishing and perfecting modern logistics information system,improving logistics infrastructure construction,cultivating and introducing modern logistics professionals and formulating logistics coordination development plan for the development plan of logistics industry in Henan Province.