An Overview of Technologies for Individual Trip History Collection: Mobility Decision Science Pillar SMART Mobility Consortium

Abstract

Department of Energy (DOE) reports produced after 1991 and a growing number of pre-1991 documents are available free via www.OSTI.gov.

Understanding Google Location History as a Tool for Travel Diary Data Acquisition

“…Similarly, while researchers have made progress transcribing passive LBS data into trips (origin-destination), it requires extensive processing and inferences to be made (43). The same can be said of raw CDR datasets, as they are not designed for modeling purposes and considerable processing is typically required: for example, to address issues of locational uncertainty and oscillation or to distill travel diary data (32,35,38,39). Travel Survey Applications.…”

“…The passive nature of this data collection has the potential to overcome reporting error found in travel diaries and household surveys ( 2 ), and to avoid compliance issues associated with GPS studies ( 2 , 19 ). For instance, smartphones typically serve a key role in participants’ daily routine, making them less likely to be forgotten and thus less vulnerable to missing data, a strength for studies with longer durations ( 32 ). Additionally, the rise of cell-phone-only households and young adults’ tendency to exclusively use mobile phones, adds promise to smartphone-based methods as a way to increase representation of younger populations (e.g., Patterson and Fitzsimmons) ( 18 , 25 ).…”

“…Despite low collection costs and broad temporal and spatial coverage, the use of passive mobile phone data poses numerous challenges. First, as researchers using CDR data are unable to interact with mobile users, they cannot obtain the important socio-economic and demographic data found in household surveys—a lack of depth also inherent to LBS data ( 32 , 39 ). Second, careful analysis is required to interpret sample bias since mobile phone users are unlikely to represent a random sample of the population, especially among LBS data, as purpose-built applications often cater to a specific group (e.g., fitness) ( 39 , 48 ).…”

“…From the outset, development and maintenance of these custom applications require specific expertise that may not be readily available to researchers ( 18 ). For example, the technical challenge of achieving compatibility across an ever-growing combination of devices and mobile OSs ( 32 ). This extends to the “backend” where processes for properly receiving and storing personal data must be developed with special attention to user privacy and data security ( 18 ).…”

“…Additionally, it is critical that applications are designed with consideration for battery life and provide a coherent and unobtrusive user experience ( 18 ). In regard to participant burden, the user must ensure that an application is properly installed, configured, and in operation for the duration of the study to avoid under-reporting trips ( 32 ). Lastly, the challenge of recruiting or popularizing travel survey applications generally results in small sample sizes ( 48 ).…”