Application of Support Vector Machines for Container Throughput Forecast

To counter the limitation of container throughput forecasting method adopted commonly at present, support vector machines(SVM)regression method is used for forecasting port throughput. This paper provides the steps of throughput forecast and determination method of relevant parameters, based on which, the application process of the method is presented in detail combining container throughput forecast of Shanghai Port as an example. The result shows that this method is reasonable and effective, and it offers a new solution for nonlinear system forecast such as container throughput, etc.