Executive Summary

Motorists circling or cruising for on-street parking that is free or priced below market equilibrium can contribute to additional congestion, air pollution, time wasted, driver frustration, and a potential loss of economic competitiveness at destinations where parking is hard to find and where alternative access modes are limited. With increased sensitivity to the need for curb management, there is a need to better understand the prevalence of cruising for parking.

Strategies to quantify cruising have evolved from intercept surveys to deployment of wireless technology sensors. Intercept surveys are effective at understanding the proportion of traffic that is looking for parking (which may not imply excess travel),¹ while the wireless technology sensor deployment has been successful in distinguishing vehicles within the traffic stream that are circling for parking; both can be applied in very limited geographies.

The methodology and tool presented in this report extends previous work that relied on global positioning system (GPS) breadcrumb data to overcome the physical limitations of other research methods. Big data allows for a more comprehensive assessment of the extent and location of excess parking search.^{2, 3} The method also quantifies the measure of most direct policy interest—excess travel from cruising, rather than the proportion of drivers searching for parking. The tool developed to generate reliable estimates of cruising (called Cruise Detector within this report) is a computer program that takes location data harvested from smartphones and sorts out which series of data points represent trips and which do not. Identified trips are then matched to a street network and compared against a shortest path that might have been taken. A trip that exceeds an available shortest path by a given threshold is assumed to include excess travel, most likely, due to parking search.

The methodology and tool provide a data-driven way to identify the locations and times of day where cruising is most prevalent. They can be used by municipalities and other interested parties to understand cruising for parking and the effects of policy interventions on parking search behaviors in order to develop appropriate responses. The methodology and tool can be applied by a skilled geographic information system (GIS) analyst with some familiarity with programming languages. Potential computing resources are also described within the report.

The research team applied the tool and completed case analyses for four U.S. cities: Washington, DC; Atlanta, Georgia; Chicago, Illinois; and Seattle, Washington.

The cases illustrate a range of applications, such as identifying cruising hot spots by both time of day and location and assessing policy impacts:

• Washington, DC. The focus in this case study is cruising for parking across Washington, DC. Special attention is focused on three sports stadiums, and when the stadiums have and do not have events planned. The analysis also emphasizes cruising in the areas around Metrorail stations. In lieu of a specific policy change that would lead to a before/after analysis, the findings here are cross-sectional, illustrating different cruising patterns for different urban forms: specifically proximate to Metrorail versus beyond the Metrorail catchment areas.
• Atlanta, Georgia. The research team explored the geography of cruising across two different time periods. To the extent possible, the research team looked at the differences in single use versus mixed use areas and contrasted metered areas with those that allow free parking. The data covered the areas of Atlanta located in Fulton County.
• Chicago, Illinois. This case study allowed the research team to showcase analyses that could be done with the raw location data. Raw location data were acquired for June 2019, 2020, and 2021. Unfortunately, the data for June 2021 were of insufficient quality to include in the final analysis. The analysis proceeds with a focus on the 2 years for which data are of sufficient quality.
• Seattle, Washington. Seattle has a performance pricing policy for its metered parking streets. Using annual surveys to estimate vehicle occupancy, the Seattle Department of Transportation (SDOT) raises, lowers, or leaves the same the parking meter prices in order to hit an occupancy/vacancy target. The research team examined cruising before and after price changes to the city’s metered parking. The research team also examined cruising in the 2 weeks before and after SDOT temporarily suspended all parking meters to ease the burden on essential travel during the worst of the COVID-19 pandemic.

The report provides analyses results for each city, as well as overall observations considering all cities’ analyses. The highest rates of cruising were found in Seattle and Chicago where 7.3 and 6.8 percent of trips, respectively, showed some portion as cruising.

Through the analyses of the cities, the research team concluded the following:

• Across all the cities in this analysis, the level of cruising is consistent, even when using different data sources. The estimates in this report are also comparable to earlier work using a similar methodology.
• The consistency, even at times of day or in places where parking is readily available, suggests that many trips identified as cruising may not be people searching for parking, but rather people taking a longer route for other reasons. Thus, estimates of cruising may overstate the parking problem.
• Consistency in the cruising estimates also point to an equilibrium level of cruising. Other work indicates that where parking is perceived to be scarce, drivers will often park short of their destination, taking the first space they find.⁴ Where parking is readily available, drivers may be more selective about their choice of a parking space. An analogy is how roadway congestion reaches an equilibrium as some users switch modes or departure times based on their tolerance for traffic delay.
• In general, cruising is a localized problem. The tool can identify hot spots, but even in these hot spots, the average time spent cruising is brief, and cruising only impacts a relatively small percentage of trips.

The report documents lessons learned relating to data quality, tool implementation, and the analyses results. For example, both third-party processed data and raw location data have their own advantages and disadvantages. Therefore, users should carefully assess options for data acquisition, considering factors such as budget and the degree of flexibility desired to conduct additional analysis to find new results beyond what was initially sought. Data quality can vary greatly, and it is recommended to obtain samples of data to assess the resolution and quality. Information on how the data are collected may help the analyst assess potential biases.