Study Of Port Cargo Throughput Forecasting Based On GPM (1,1)

The port cargo throughput is influenced by economy, politics, culture, technology and natural environment,and so on, then, the relationship among of various factors forms a complicated nonlinear system, but the change also has internal regularity. In order to choose appropriate port cargo throughput prediction model,it's important to acquisite variation and inner link of these factors, and master these factors change regularity. With national rapid development, each port cargo throughput changes by the obvious increase trend and nonlinear variation trend, and gray theory has certain advantages in this prediction of growth-type nonlinear time series applications, but there are prediction accuracy of problems.Therefore, this paper try to achieve the purpose of highly forecasting accuracy by improving traditional GPM (1, 1) model.On the other hand, although the grey theory in port throughput prediction model is more, but the application of gray theory model was developed in power was slower and less. Therefore, this paper proposed improved GM (1, 1) model combining metabolism method to evaluate the feasibility and effectiveness of the prediction model.After the model has done, based on port cargo throughput of Guangzhou port from 1998 to 2009, the paper simulate , analyse, compare and verify for the model ,confirming that the prediction model is feasible and efficient. Meanwhile , the port cargo throughput of Guangzhou port from 2010 to 2015 is forecasted, and it will provide scientific reference data for Guangzhou port development and planning.