This study demonstrates a segment-based approach to integrate GIS and GPS data to address questions about the connections between the built environment and travel behaviors. Methods and challenges of GPS/GIS integration are discussed, and an application integrating GPS naturalistic driving data from Southeast Michigan together with GIS data from several sources is demonstrated. The integrated dataset is used to explore connections between the built environment and driving behavior, specifically between business concentration, driving speed, vehicle stops and rear-end crashes. Driving speed, an important determinant of driver behavior linked to traffic safety, is found to be inversely related to business concentration, a pattern that does not vary by time of day. Rear-end crashes are found to increase with vehicle stops which increase with business concentration. This demonstration showed that fusing GPS and GPS data provides spatial intelligence which can be used to address planning, traffic safety, and transportation related issues.
TopIntroduction
The relationship between travel and the built environment remains a topic of interest and importance in many fields including transportation, urban planning, and public health. The built environment as described by land use patterns and road configurations, is considered to be one of the key determinants of travel behaviors and, by extension, travel safety, energy consumption and pollutant emissions (Banister, Watson & Wood, 1997; Bhat & Guo, 2007; Cervero & Kockelman, 1997; Ewing, Bartholomew, Winkelman, Walters & Chen, 2008; Ewing & Dumbaugh, 2009; Frank, James, Terry, James & et al., 2006; Handy, 1992; Hanson & Schwab, 1986; Maat & Timmermans, 2009). Geographic information systems (GIS) play a key role in advancing our knowledge in the relationships between the built environment and travel outcomes. The increased availability of GIS-based spatial data and advances in GIS tools allow for a more comprehensive development of built environment measures, which could quantify the density, diversity, and design of the built environment (Cervero & Kockelman, 1997; Galster et al., 2001; Krizek, 2003; Song, 2004; Witten, Pearce & Day, 2011). Studies incorporating the newly-developed GIS measures show significant relationships between the built environment and travel outcomes (Cervero & Kockelman, 1997; Frank, et al., 2006).
Along with advancements in GIS measures of the built environment, there has been widespread use of global positioning systems (GPS) for scientific, commercial, navigation, tracking, and surveillance purposes, generating much GPS data on people’s travel. This presents opportunities for transportation professionals to monitor and quantify travel behavior in ways that have never been done before. Researchers and practitioners in the transportation field rely on models that are based on data on where, when, and how people travel. Currently, data are collected through traffic volume counts at specific locations and through household travel surveys that rely on respondents completing travel diaries for one or two days. Traffic counts by themselves do not reveal much about travel behavior, and travel surveys are costly, time-consuming, and do not record the specific routes on which people travel. Compared to the traditional trip diary data, GPS data provides highly accurate spatio temporal information including travel routes. Furthermore, GPS technology is capable of tracking human or vehicle movements with a finer level of resolution and during much longer periods, even while imposing fewer burdens on survey respondents (Grengs, Wang & Kostyniuk, 2008; Wolf, 2001; Wolf,. Hallmark, Oliveira, Guensler & Sarasua, 1999).