“Big Data” is a valuable resource for the future of mobility analysis. But could unmeasured biases in the data be influencing transportation infrastructure spending? 

PROJECT OVERVIEW

Numerous studies have now used emerging big data (e.g., mobile phone data, app-based data, social media data) for mobility analysis, i.e., understanding how individuals move in time and space. Information derived from such data tells where Americans live, work and go about other activities; such information is critical as they are the basis for hundreds of billions’ investment for the nation’s transportation infrastructures. All such data, though having an enormous size, are unrepresentative(biased) due to its self-selection nature—users of such datasets self-select certain kinds of mobile services(e.g., cellular, social media, apps). Only a few studies have attempted to quantify the associated biases and uncertainties and there exists little to no knowledge about the extent of the biases and uncertainties involved and what caused them. This research proposes a whole-community research effort to answer two questions: 1) how large are the biases and uncertainties associated with the critical mobility metrics (e.g., trip rates) and how they vary for different geographical areas and population segments; and 2) to what extent the biases and uncertainties associated with those metrics are affected by a number of factors, e.g., data characteristics, the modeling techniques used, and geographical differences

PROJECT GOALS

Thrust 1 engages stakeholders and the entire research community to develop and issue a community call calling for mobility labs around the world to submit a set of critical mobility metrics, using their own data and methods.

In Thrust 2, the PI team will conduct a meta-analysis of the information submitted from participating labs. This meta-analysis will answer those two questions.

Thrust 3 involves a virtual meeting with all participating labs, followed by a final in-person community workshop focusing on results summarization and dissemination.

Funded by the National Science Foundation (NSF) Civil Infrastructure Systems (CIS) Program

Please check back for project updates