Identifying trips and activities and their characteristics from GPS raw data without further information

Abstract

In recent years, studies based on GPS records have become more and more important due to their manifold advantages compared to classic survey methods, such as paper diaries or telephone interviews. However, without additional information, a lot of post-processing work is required to derive data that can be used for analysis and model estimation. These post- processing procedures are still an ongoing research issue. Recently, a couple of new ideas have been published, but the key research questions remain the same. This paper focuses on the description of a post-processing procedure that is able to determine trips and activities, including several characteristics, such as modes and routes from GPS raw data without any further information. It is applied to GPS records collected in the Swiss cities of Zurich, Winterthur and Geneva with 4882 participants, each of which carried an on-person GPS-receiver for 6.65 days on average. The data outcome is compared to the Swiss Micro- census 2005 to prove that it is ready for further applications, such as discrete choice model estimations.