港口货物吞吐量双层预测模型研究

Research on Double Forecasting Model of Port Cargo Throughput

在得出港口货物吞吐量概率分布的基础上,利用条件数学期望提出了港口货物吞吐量的双层预测模型。由于货物总吞吐量的变化与到达港口的货运船数目以及装卸设备的工作效率有密切关系,构造一个关于到达港口的货运船数目以及装卸设备的工作能力组合而成的复合变量,货物总吞吐量是这些复合变量所表示的货物装卸量的和。应用全概率理论,得到货物总吞吐量的概率分布。在此基础上,将未来货物总吞吐量看作已完成吞吐量的条件期望,利用增长函数得出港口货物吞吐量的预测模型。以山东地区某港口的货物吞吐量变化规律进行了案例分析。理论分析和案例分析均表明该模型的计算误差要比其它方法小。

A double forecasting model is proposed by using conditional mathematical expectation,based on probability distribution of port cargo throughput.Compound variable,being composed of the number of arrived cargo ships at the port and the operation capacity of handling equipment,is established,port cargo throughput is then the sum of all compound variables.Probability distribution of cargo throughput is obtained by using throughout probability theory.On this basis,future port cargo throughput can be taken as the conditional mathematic expectation of realized throughput,a forecasting model is proposed by applying growing function.A case study of cargo throughput of a port in Shandong province is carried out.Both theoretical analysis and case study show that this model based on conditional expectation has smaller errors of forecasting port cargo throughput than other models.