EXPLORING THE USE OF PASSIVE GPS DEVICES TO MEASURE TRAVEL

摘要

The high completion obtained by the prompted recall method with route maps suggest that this technique is successful and can be used for future surveys. The accuracy of the travel time and distance data is clearly demonstrated in the pre-pilot survey, as well as data on routing, origin and destination. The prompted recall was able to complete all of the desired data with ease, even after a lapse of up to 14 days from when the travel was undertaken. It is not expected that this type of survey would be conducted on several thousand households in a metropolitan area. It holds out promise as a means of correcting data obtained from a standard household survey. While no research has been done on this aspect, it appears likely that further research would produce methods whereby travel times and distances could be corrected, and better imputation could be performed for the origins and destinations that people are unable to identify sufficiently for geocoding in conventional surveys. In addition, there are clear potentials to analyse route choices, congestion, speed profiles, and emissions-related statistics. The GPS protocol holds out considerable promise of providing a relatively inexpensive and easily-conducted survey that can provide accurate geographic and time data on household members' travel. A high level of accuracy in the data collection is achievable and the device is simple and easy for subjects to use. The prompted recall survey appears to hold considerable promise, and it appears to be worthwhile to examine its sensitivity to elapsed time between the travel and the interview.