Shanghai Inland Transport Survey System Problem

There are ninety-three registration stations in shanghai freshwater, most of which register vessels pass in and out manually. Because of large amount of data registered, Shanghai Navigation Administration Department doesn't have enough persons and appropriate method to report navigation statistics index to superior department, so at the moment three days continual investigations are taken every month. Because all registration stations are investigated and the data for each station is drawn at random for three days, the workload is very large. Although large workload, the final conclusion usage is uncertain for gross estimate error can't be calculated caused by investigation not being strictly random.At moment, leaders of Shanghai Navigation Administration Department plan to equip all ninety-three registration stations with computers to investigate continually in twenty-four hours. But this method needs higher criterion for manpower, resources and finance. First, twenty-four hours continual investigation will increase manpower. Second, it needs more qualified persons, and they must know both statistics and computer. Finally, great and standing financial power is needed for computer original purchase and further maintenance and renewal because of the great mount of registration.? According to historical data from Shanghai Navigation Administration Department, the paper uses theories from survey sampling and multiple statistics to design sampling method and calculate sample size in order to change present condition of investigation all stations and find more economical computer registration station allocation scheme.This paper contains three parts: preface, text and conclusion. The text includes four chapters, they are the mains of the whole paper.In preface part, the paper points out research content and method based on realistic demand and present condition analysis. First chapter mainly introduces basic theories and characteristics of survey sampling, which include basic concept, basic sampling method and characteristics of survey sampling. Second chaptermainly makes simple researches of organization methods of statistics survey, which include simple research of statistics survey population and stratification of survey sampling by cluster analysis. Third chapter mainly introduces sampling scheme design and calculation of sample size, among which catalogue sampling used in the scheme is introduced by detail. Fourth chapter mainly introduces another consideration of computer registration station finding, that is, find computer registration station by regression estimate. In the conclusion part of the paper, the proposal scheme for Shanghai computer registration station allocation and water shipment statistics survey is given and research result is evaluated.