Every day, drivers across towns and cities rely on navigation apps to estimate travel times. However, these apps often fail to consider the additional time required to find parking upon reaching the destination. This underestimation, besides causing frustration for drivers, exacerbates congestion and increases emissions as motorists cruise around in search of parking spots. It also discourages people from considering faster alternatives like mass transit. However, a new system developed by MIT researchers aims to address this issue.
A Smarter Approach to Parking
The MIT research team has developed a system that identifies parking lots offering the best balance of proximity to the destination and the likelihood of parking availability. Unlike conventional navigation systems, this method directs users to the ideal parking area instead of the final destination. In simulated tests with real-world traffic data from Seattle, the technique achieved time savings of up to 66 percent in the most congested settings. This translates to a reduction in travel time of about 35 minutes for a motorist, compared to waiting for a spot to open in the nearest parking lot.
Although the system hasn’t been designed for real-world application yet, the demonstrations suggest the viability of this approach and provide insights into its potential implementation. As explained by Cameron Hickert, an MIT graduate student and the lead author of the research paper, “This frustration is real and felt by a lot of people. Systematically underestimating these drive times prevents people from making informed choices. It makes it that much harder for people to make shifts to public transit, bikes, or alternative forms of transportation.”
Understanding the Probability-Aware Approach
The researchers developed a probability-aware approach that considers all possible public parking lots near a destination, the distance to drive there from a point of origin, the distance to walk from each lot to the destination, and the likelihood of parking success. This approach, based on dynamic programming, calculates the best route for the user by working backward from good outcomes.
The method also factors in scenarios where a user arrives at the ideal parking lot but can’t find a space. It takes into account the distance to other parking lots and their respective probabilities of parking success. As Hickert explains, “If there are several lots nearby that have slightly lower probabilities of success, but are very close to each other, it might be a smarter play to drive there rather than going to the higher-probability lot and hoping to find an opening. Our framework can account for that.”
Factoring in Other Drivers
The system also incorporates the actions of other drivers, as they can affect the user’s probability of parking success. For instance, another driver may arrive at the user’s ideal lot first and take the last parking spot. There might be spillover effects from motorists parking in different lots, which lower the user’s chances of success. “With our framework, we show how you can model all those scenarios in a very clean and principled manner,” says Hickert.
Leveraging Crowdsourced Parking Data
For the availability of parking data, the researchers propose using crowdsourced information. Users could indicate available parking through an app, or data could be gathered by tracking the number of vehicles circling to find parking or those entering a lot and exiting after being unsuccessful. In the future, autonomous vehicles could report open parking spots they drive by.
The researchers evaluated their system using real-world traffic data from Seattle, simulating different times of day in a congested urban setting and a suburban area. They found that their approach cut total travel time by about 60 percent compared to waiting for a spot to open, and by about 20 percent compared to a strategy of continually driving to the next closest parking lot. According to their findings, crowdsourced observations of parking availability would have an error rate of only about 7 percent, indicating its effectiveness in gathering parking probability data.
Looking ahead, the researchers plan to conduct larger studies using real-time route information in an entire city. They also aim to explore additional sources for parking availability data, such as satellite images, and to estimate potential emissions reductions. As Cathy Wu, the senior author of the study and a professor at MIT, concludes, “Transportation systems are so large and complex that they are really hard to change. What we look for, and what we found with this approach, is small changes that can have a big impact to help people make better choices, reduce congestion, and reduce emissions.”
The research was supported, in part, by Cintra, the MIT Energy Initiative, and the National Science Foundation. Read more about it Here.
