GIS-based episode reconstruction toolkit (GERT): A transferable, modular, and scalable framework for automated extraction of activity episodes from GPS data

The widespread use of global positioning systems (GPS) has prompted transportation researchers to develop tools and methods to extract information from person-based GPS data. However, most of these procedures suffer from specific data requirements and complexity that limit their transferability. Further, they have a limited set of modules to extract all necessary information (e.g., route attributes), and were not specifically designed to handle huge GPS datasets. To deal effectively with these problems, this paper presents a framework based on three design principles (transferability, modularity, and scalability), along with the geographic information system (GIS)-based episode reconstruction toolkit (GERT) based on this framework, for automated extraction of activity episodes from GPS data. About 26,000 episodes were automatically reconstructed using GERT from 47.3 million GPS points. A comparison of the episode and duration distributions reveal similar patterns between time-use diary and GPS episodes, a similarity that confirms that GERT's modules work properly in reconstructing episodes from GPS data. GERT's overall performance suggests potential because of its scalability - GERT can scale up to large GPS data; modularity - GERT has a complete set of tools to support activity analyses and route choice model estimations; and transferability - GERT's reliance on generic variables (latitude, longitude, time) makes it applicable to other places. Overall, GERT's modules provide transportation researchers with rich datasets (stop and travel episodes, activity locations, travel segments, route choice sets, route attributes) for improving the understanding of activity/travel patterns in general and route choice decisions in particular. (C) 2017 Hong Kong Society for Transportation Studies. Published by Elsevier Ltd. All rights reserved.