01553668

Component

In this paper, the authors analyze data from a large-scale survey, the 2003 Tokyo Metropolitan Freight Survey (TMFS), to examine some of the sampling issues associated with urban freight surveys. The TMFS data indicate that the variables commonly obtained from freight surveys, such as tons shipped and truck trips generated at a facility level, have highly skewed population distributions, making it difficult to obtain accurate estimates when the sample size is modest. Using bootstrap method, the authors compared the performances of stratified sampling against simple random sampling. The analysis indicates that, under the ideal condition, stratified sampling with Neyman allocation can improve the accuracies of the estimates. However, two-phase sampling, which is often used in practice to obtain the information that is necessary to implement the Neyman allocation, tends to produce poor results. The findings presented in this study underscore the importance of sample size in freight surveys and the efforts to capture mega shippers in freight surveys.

This paper was sponsored by TRB committee ABJ90 Freight Transportation Data.

01550057

15-4718

English

16p

2015

Transportation Research Board 94th Annual Meeting

Digital/other

Figures; References; Tables

Accuracy; Freight transportation; Random sampling; Surveys; Urban areas

Bootstrap analysis; Freight data; Mega-carriers (Shipping); Sample size; Stratified sampling method

Tokyo Metropolitan Area

Data and Information Technology; Freight Transportation; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2015 Paper #15-4718

TRIS, TRB, ATRI

Dec 30 2014 1:34PM