Data from roadside monitoring equipment, cell phones, license plate readers, and similar technology are very useful for transportation planning and traffic engineering. New transit fare systems also collect individualized data that are similarly useful. Data generated for a single purpose by personal technology increasingly can be useful for transportation planning purposes. These data, however, often include personally identifiable information and may be subject to privacy concerns and/or legal restrictions.
Large geographies are a very simple way of making data anonymous but this significantly reduces the utility of the data. It is therefore important to maintain a high level of geographic granularity while respecting privacy concerns. Methods to make or keep data anonymous may provide a way of continuing to collect needed data without violating privacy concerns or restrictions The 2007 National Academies Press report, “Putting People on the Map: Protecting Confidentiality with Linked Social-Data” (https://books.nap.edu/openbook.php?record_id=11865) provides additional background information for this research.
The objectives of this research were to:
1. provide an inventory of methods used by transportation agencies as well as other public organizations and private entities to make data anonymous, or collect it anonymously in the first place;
2. identify the features that are of concern from a privacy perspective and that would be of interest to transportation agencies;
3. characterize the tradeoffs when certain features are anonymized while preserving the usability of the data, e.g. spatial granularity, as well as their statistical validity;
4. assess the applicability of the methods for transportation planning and traffic engineering purposes: and
5. identify future research and development needed to provide effective methods for making data used for transportation planning and traffic engineering anonymous.
The Final report is available HERE.