Research Of Sampling Technology And Application On Railway Passenger Flow Investigation

It is well-known that China has been developing its railway construction. Hence, under this overall background, the efficient and accurate information about passengers is beyond all doubt needed to improving railway transportation enterprise's social and economic effect. The approach for obtaining this information is by doing passenger flow investigation.This thesis aims to point out an integrated sampling technique, to conduct the sample frequence and the sample population in passenger flow investigation, as well as the assessment about the perfection degree of the information, so as to improve the quality of investigation by three aspects-"accurate", "complete", and "practical". Further more, the high-quality data is also supposed to provide reference frame for decision making.In sample freqence aspect, this thesis introduces a group sampling method, which is led by analyzing the discipline on passenger volume-time variation. By the minimum layer variance, the larger internal differences can be divided into smaller sub-total internal differences, so that the sampling time and interval has been acquired.On the premise that the sample frequence is given, this thesis raises a method which is based on small-sample approaching method, by doing sample analysis on each factors. Then, this thesis finds the relevance between the established sample and dispersion degree, providing basis for fixing the minimum sample population.In the assessment of perfection degree of infromation, this thesis is inspired by comentropy, and then raises an evaluation methodology and critism. Using the table and chart shown in this paper, the raised evaluation methodology provides foundations for questionaire designing.At the end, the sampling mehtod in this thesis is applied in the investigation of Cheng-Guan railway survey. By analysising with singal factor, some regular pattern about passenger flow structre is obtained. On this basis, K-means clustering algorithm is used to analyze the conclusion of the passenger survey, the Cheng-Guan railway market is segemented into three sub-market segments. Finally, some marketing advice is given pointedly in designing ticket.