Lead Presenter Biography

Gargi Singh is an Assistant Research Scientist at Texas A&M Transportation Institute. Prior to this, she went to Texas A&M University, where she received her master's degree in Urban Planning. She is a data enthusiast and a self-taught programmer with six years of experience working with passive mobility data. Her research focuses on using such datasets to solve transportation problems while deep diving into data to understand and account for bias.

Co-Authors

Presentation Description

The Texas Department of Transportation (TxDOT) has been conducting Special Generator surveys in conjunction with Workplace surveys as part of its Texas Travel Survey Program. Special generators are developments or activities that are unique to a region and that attract and produce significantly more trips than would be indicated by workplace employment, square footage, or land area. These typically include regional hospitals, major universities, community colleges, regional shopping malls, recreational facilities, airports, military bases, and other land use developments. However, due to their large size and trip-generating capacity, special generators are often difficult and costly to survey. They can include many locations for conducting intercept interviews and vehicular and/or pedestrian counts. With the high costs involved in implementing the special generator surveys, as well as the complications in logistics and the diminishing survey participation levels and reluctance to interviews, TxDOT has considered using passive data sources to supplement and potentially replace these surveys. SafeGraph's Point-of-Interest (POI) data was evaluated to assess its viability for use in these surveys.

Safegraph provides visitor (foot traffic) counts for business listings by North American Industry Classification (NAICS) code and by specific business locations. This information is primarily composed of data aggregated from a variety of smartphone applications. The data is further processed by vendors to exclude duplicates and then associated with building footprints to infer the site or building footprint of individual establishments or geographic areas of interest.

As part of the study, SafeGraph’s POI data was purchased for the entire state of Texas and compared with traditionally collected survey information to evaluate the POI data’s suitability in obtaining special generator data and understand its penetration rate at the site level. Total traffic to and from a workplace or SG survey is the most important data element collected in the survey since it provides total trip generation for the site and serves as the means to expand survey results. The study was conducted to assess the use of POI passive data as an alternative means for collecting traffic counts at survey sites, in lieu of conducting actual counts at the site. In addition to that, the data was scrutinized across multiple factors, such as time of the day pattern, visitor counts, building footprint, visitor home location, etc. The results of the study highlighted the strengths and weaknesses of both passive POI data and travel survey data. 

Presentation File