Data-Driven Parking in Seattle's Belltown North Neighborhood
business & systems analysis
data science & visualization
An average driver in Seattle spends 58 hours a year looking for parking. To reduce this burden, city officials must first understand on-street parking occupancy and parking behaviors. The challenge is that pay station transaction data doesn’t reflect actual parking occupancy throughout the year. To reconcile this problem, we leveraged publicly available information assets to build statistical models to predict paid parking occupancy in Belltown North. Our machine learning model, analysis of factors related to occupancy, and documented process is moving the needle toward a citywide system of policies and driver tools that streamline the parking experience.