The Combined Model Of Express Demand Forecast And Empirical Study

As world trade and domestic trade become more active, express industry plays a more and more important role in modern social economic activities, driving the development of other industries, and then gets great attention from government. Relevant policies were issued successively to guide the development of express industry steadily. Delivery forecast is the basis of express delivery planning. In view of this,this paper takes express demand forecast as the research object, aiming to establish proper express demand forecasting model with its practical value.This paper based on the available literatures, in combination with the practical situation of our country, firstly,analyzes the characteristics,the influence factors of express delivery demand and the steps of forecast, Secondly, selects eight influence-related indicators considering the data availability such as GDP?TRSCG and so on, using grey relational analysis to testify the relationships of express delivery demand and eight influence-related indicators, proving that total export-import volume and internet access are two important indicators from 2006 to now. Thirdly, in order to achieve the purpose of easy operation, high accuracy and strong applicability, chooses GM model, trend extrapolation and multivariate linear regression model from the existing time sequence prediction models and causality prediction models, and builds combined forecasting model based on shapley value distribution principle. Lastly, collects the data of the influence-related indicators of Sichuan from 2006 to 2016, forecasts the delivery demand of Sichuan province and tests. The result shows that the combined forecasting model is suitable for medium and short period express demand forecast with higher accuracy and meets the actual demand and the express demand of Sichuan will still increase in the next few years.