Demand-Responsive Pricing on the Cheap: Estimating Parking Occupancy with Meter Payment Data

Abstract

The SFpark pilot by the San Francisco Municipal Transportation Agency in California was the first large-scale test of demand-responsive parking pricing in a major city. Several evaluations of the pilot showed that the project yielded substantial benefits. However, measuring parking occupancy is critical to implementing demand-responsive pricing. San Francisco relied on wireless in-ground parking sensors to measure parking occupancy for the SFpark pilot, but those sensors met the end of their useful lives and were deactivated. Parking sensors are still a nascent and costly technology that presents a great deal of risk to cities. Yet many cities, including San Francisco, are adopting new parking meters that make meter payment data widely available. Using sensor and meter data from the SFpark pilot, the agency developed a sensor-independent rate adjustment model that estimated parking occupancy by using meter payment data. Although not everyone pays the meter when they park, the model can reliably estimate occupancy to support demand-responsive pricing. This capability allows San Francisco to continue its SFpark program and lays the foundation for other cities to implement demand-responsive pricing and promote the benefits of better parking policy more widely.