A Novel Grey Seasonal Prediction Model for Container Throughput Forecasting.

Containerization is regarded as an important driver of globalization and international trade, and it also drives the development of global ports. Seasonal container throughput prediction is crucial for planning and operation by port authorities, and for the strategies formulated by logistics companies. To accurately predict the seasonal fluctuations in port container throughput, we propose a novel grey seasonal model called, FNDGSM(1,1). The proposed model involves lime item, cycle Hausdorff fractional accumulating generation, and seasonal dummy variables. The particle swarm optimization algorithm is used to obtain the optimized parameters. Experimental results demonstrate that the proposed seasonal grey prediction model performs significantly better than other prediction models with quarterly container throughput data. [ABSTRACT FROM AUTHOR]