Harbor transportation volume prediction method and system

Abstract

The invention discloses a harbor transportation volume prediction method and a system. The prediction method comprises steps of after carrying out preprocessing like correlation analysis and dimensionality deduction on related factors of the harbor month transportation volume, accumulating the harbor month transportation volume and influence factor data sequences with the gray prediction method; carrying out parameter optimization on a support vector machine by use of the indetermination knowledge particle swarm optimization algorithm so as to obtain an SVM model; carrying out prediction so as to obtain prediction results based on the model; and at last obtaining the prediction results after correction of the harbor month transportation volume via continuous subtraction reduction. According to the invention, by combining the gray prediction method, the particle swarm optimization algorithm and the SVM model, the transportation volume is predicted, advantages of all models or methods are given into full play and enhanced while disadvantages are avoided, thereby achieving precise prediction for transportation volume of all kinds of goods of the harbor in a simple, efficiency and low cost manner, and filling the gap in application of the harbor transportation volume field.

Description

technical field [0001] The invention relates to the field of forecasting volume of transport, and more particularly to a method and system for forecasting volume of port transport. Background technique [0002] As an important economic material of the country, bulk cargo (including bulk cargo and piece cargo) occupies a very important position in the shipping market. At present, my country's water transportation is still dominated by bulk cargo. According to the Statistical Bulletin on the Development of Highway and Waterway Transportation Industry in 2011, among the various types of cargo throughput in the national ports, dry bulk cargo accounted for 58.3%, liquid bulk cargo accounted for 9.1%, and general cargo accounted for 10.1%. The proportion is 17.7%. Compared with container ports, bulk cargo ports have more complex loading and unloading processes and a wider range of operating cargoes. Container ports have a higher degree of specialization, and the cargo loading and...

AI Technical Summary

Problems solved by technology

[0004] At the research level, among domestic scholars, Zhang Shuai of Dalian Maritime University predicted the throughput of Jinzhou Port in 2013, and obtained the annual throughput values ​​of major bulk cargoes such as oil, coal, steel and grain. This prediction method is relatively rough , has no guiding significance for specific production practices, and cannot be used as a basis for port production scheduling

Song Xin of Beijing Jiaotong University based on a number of improved BP neural networks to predict the monthly coal transportation volume of bulk cargo ports, and obtained meaningful conclusions. However, the prediction methods based on neural networks generally suffer from high network training costs and a large amount of calculation. , can not guarantee the convergence and poor generalization ability and other defects

Images