港口吞吐量预测影响因素筛选方法研究

Screening of influential indices for port throughput forecasting

物流预测指标的筛选是物流需求预测工作的首要步骤。对港口吞吐量及其影响因素进行了分析,提出了预测指标的筛选原则和筛选方法,并以广西北部湾港口为例,利用2000—2008年的统计数据,采用灰色关联分析法和DPS数据处理软件工具求出各因素与港口吞吐量以及各因素之间的灰色关联度,最后根据筛选原则选出一组相关性强、结构合理的港口吞吐量预测指标。

The screening of the logistics forecasting indices is the first step of a logistics demand forecast.Based on the analysis of the port throughput and its influential indices,we propose a series of screening principles and methods.Taking the ports in the North Gulf of Guangxi as an example,applying the Grey Relational Analysis method and DPS（data processing software） tool,we find the correlations between the port throughput in the North Gulf of Guangxi and related forecast indicators based on the data from 2000 to 2008.Finally,a group of well-constructed indices highly correlated are chosen as the final port throughput forecasting indicators.