Logistics Park's Logistics Volume Forecast Based On Intelligent Forecasting Model And Its Empirical Study

Logistics Park is important logistics node enterprise and logistics infrastructure. Its volume forecast is the essential link in plan and construction of Logistics Park. Logistic volume forecast provides not only the quantitative basis for planning logistics park's location,construction scale and infrastructure scale, but also scientific foundation for running logistics park's operation, management decision and future development. On the background of logistic industrial developing fast and Logistics Park construction rising, this paper studies its volume prediction.A comprehensive literature review on domestic and international related study and theory on Logistics Park and logistics volume prediction is conducted. Based on understanding and analysis on the meaning of logistics volume, the idea and method of logistics volume measurement are provided. Then combined with the characteristics of the logistics volume of logistics park, qualitative and quantitative analysis, based on related theory of neural network intelligence forecast, This paper constructs the system of Logistics Park's logistics volume forecast with three intelligent forecasting models. They are the time series neural network, regression neural network and nonlinear combination network. The three intelligent forecasting models will be achieved by using BP algorithm. On the empirical study of Loudi Logistics Park planning and construction, According to the idea and processes of logistics volume prediction of Logistics Park provided by this paper, the intelligent prediction function of neural network is realized by MATLAB neural network toolbox. First of all, an integrated transport volume of Loudi is forecasted by applications of intelligent logistics prediction system built by the paper. On this basis, according to the relationship between total traffic volume and logistics volume, the total logistics amount is accounted by a certain accounting standard. Then the logistics volume distribution of Logistics Park is carried out and final logistics volume of Loudi Logistics Park can be achieved. Through contrast analysis on the effect of the forecast models, the feasibility and superiority of the neural network on prediction system are obtained. Of which, non-linear combination of intelligent predictive has a higher prediction accuracy than a single forecasting model.