CEO & Co-Founder
January 19, 2021
For most people looking back at 2020, the parking industry wasn’t even a blip on the radar. With the average citizen stuck at home and parking less than ever, demand exploded for curbside pickup, e-commerce, and on-demand delivery. Cities are suddenly dealing with a pandemic and watching parking revenues plummet, all while Instacart grows 500% YoY and DoorDash gets valued at $50B+ in the public markets.
I should clarify that I don’t think the success of those companies and industries is a bad thing - quite the opposite. I do think they’re representative of the prolonged shift in curb demand from private vehicle storage to commercial activity, and that COVID accelerated the realization that cities need to adapt if they want to maintain control.
For the sake of disclosure (and self-promotion), I’m the CEO of Vade - we use cameras and computer vision to help cities understand and manage the curb. Many of the topics within this will be relevant to what we do, but I’ll keep it informational. My goal is for anybody that reads this to learn something interesting, regardless of if they are a city parking manager, the founder of a mobility startup, or my next-door neighbor. I’ll be publishing posts like this every week, so if you enjoy please consider subscribing to the Modern Mobility newsletter! (it’s free)
The origins of today’s parking system can be traced back all the way to the 1930s when Oklahoma City deployed the first parking meter. Free and unregulated on-street parking was causing traffic and congestion, so cities wanted to charge a nominal fee for using curb space to encourage turnover. It didn’t take long for other cities to follow suit, and thus the modern parking system was born.
Turnover refers to the volume of cars coming and leaving a parking space in a given time period. There’s a handful of different math formulas that I won’t go into, all you need to know is that more turnover means less congestion (and more money).
For a while, the system worked… until now. The growth of e-commerce, rideshare, and on-demand delivery have fundamentally transformed the demand for curb space. For most of the last century, parking demand was something like 95% private vehicle storage (a.k.a. citizens parking), and 5% everything else. Today, commercial vehicles are already estimated to make up 30% of parking activity and are expected to overtake passenger vehicles within the next decade (Smart Cities Dive). Most cities knew that and had long-term plans in place - then COVID happened.
Without effective means of managing loading zones, cities are missing out on significant revenue potential. Compared to meter expiration violations, commercial vehicle violations like double parking and blocking bus/bike lanes are serious public safety issues. UPS pays over $30M annually for parking violations in NYC alone, and that's including a stipulated fine program that greatly reduces the price of citations to prevent them from using finite court resources (Gothamist). Commercial vehicles already spend up to 28% of drive time circling for parking, so parking tickets are just a cost of doing business (University of Washington). Unfortunately, the general public is left to deal with the increased congestion and dangerous driving conditions.
Effective curb management can reduce congestion, increase safety, and increase revenue for cities. With each of those benefits being more than enough to warrant a city taking action, we’re left to wonder… why haven’t they solved it?
While the history of parking is a riveting topic, it also helps explain this answer. Parking was designed for private vehicle storage where citizens are parking, leaving their cars, and spending money at local businesses. In that context, curb space is not just a cash-flowing asset, but also a public good with positive externalities.
Cities could even adapt to some commercial activity like freight trucks by designating loading zones and issuing permits, which also contributed to a vibrant downtown and thriving local economy. The new modes of mobility don’t fit within the boundaries of the modern parking system. Gig-drivers are using personal vehicles for commercial activity, rideshare vehicles often stay at the curb for less than a minute, and the massive scale and complexity behind e-commerce logistics create an incentive structure where UPS is better off paying $30M in fines than following the rules.
So what now? Like every other problem in the 21st century, the solution is technology. Over the next decade, we will see more cities adopt more technology-enabled solutions to manage the curb more efficiently, make it easier to follow the rules, and make it harder to get away with breaking the rules. If you ask me, it only makes sense for the billion dollar companies to pay for the curb space they profit from using. It’s only fair - if commercial vehicles account for 30% of parking activity, they should make up at least 30% of municipal parking revenues.
By far the most popular trend has been cities digitizing their curb inventory. It may come as a surprise to learn that most cities don’t really know how many spaces they have, which types of spaces they are, or where they are located. Well, they know in the sense that everything was documented at some point and could be dug up in theory, but they don’t have a centralized source of inventory information. This explains why commercial fleets choose to pay tickets instead of following the rules - it’s really hard, sometimes impossible, to get data about curb rules. There’s been a flurry of companies attempting to solve this by aggregating as much data as possible, and a handful of companies going one step further and offering solutions for cities to generate the data. There’s even been a few Tier-1 cities that have created, cleansed, and published their own layers of curb data like Seattle, NYC, and Austin. The fragmentation of data structures and competing data standards has created a problem in itself, but overall the trend of digitizing curb inventories took the vital step of enabling the next generation of curb management solutions.
Cities have also started to explore the dynamic allocation of loading zones to best fit the changing needs of different curb users throughout the day. There are two primary use cases: passenger pick-up/drop-off, and commercial loading. Each of these use cases contains many niche use cases like food delivery by gig-workers, freight loading, parcel delivery, rideshare, etc. Each of these primary and secondary use cases have different characteristics like average dwell time and length of curb space needed, so cities are looking to implement dynamic curb space allocation strategies that can serve many use cases throughout the day in an efficient and effective manner. Cities like Seattle, Chicago, Washington DC, and San Francisco have made impressive progress in creating more sophisticated systems of static curb space allocation, including the use of colored curbs and robust online databases accessible to commercial fleet management systems.
Lastly, cities are looking at new ways of monetizing their commercial curb zones as a tool to both generate revenue and encourage turnover. For example, NYC charges $2.50 per hour, Seattle charges $3-4 per hour, Chicago charges $14 per hour, and DC charges $2.30 per hour. Additionally, Seattle and DC offer annual permits as an alternative to the hourly cost. From a revenue generation perspective, Chicago estimated a citywide deployment would yield a 10% increase to their ~$130M in annual parking meter revenues (CMAP). As far as the non-revenue benefits: NYC saw the average dwell time reduce from 160 to just 45 minutes, with only 25% of commercial vehicles parking in the same space for more than an hour (US DoT), and DC saw a 64% reduction in double parking (GovTech). Despite encouraging data from a handful of initial pilot initiatives, we are yet to see a consensus on the future of commercial curb management policies.
Most sensor-based approaches within curb management are utilizing some form of computer vision to detect curb activity in real-time (or near real-time). Computer vision has proven to be a more accurate, more affordable, and more versatile method of data collection as compared to hockey-puck sensors, and all signs point towards it’s dominance as a data collection methodology going forward. On top of vehicle and curb space occupancy detection, computer vision models can detect the characteristics like the type and color of vehicles.
License Plate Recognition
License plate recognition (LPR) is widely considered to be the most difficult functionality due to specific and unique challenges. These include variability in the cleanliness of license plates, the format of license plates in different regions, the orientation of parked vehicles, the resolution of the camera, and the speed of execution. Due to the specialized needs, many LPR-centric vendors have emerged that provide hardware specifically designed for LPR, software specifically designed for LPR, or some combination of both.
Data Sharing and Aggregation
Integrations have always been a core component of the urban mobility ecosystem, and that trend has continued with curb management. Most new entrants place a large emphasis on API integrations (at a price), typically to reduce the switching costs of municipalities by allowing their existing enforcement, analytics, and management vendors to improve the solutions already in place with new streams of data. Many companies are developing software to aggregate different datasets in an effort to become the system of record for all curb interactions. While data aggregation is likely to remain an important way to draw meaningful insight from data, there are doubts about the efficacy of approximation within the context of a system of record.
While there are many unknowns, I do have conviction in a few predictions for what curb management and parking will look like in 2030:
1) Curbs will be allocated for various use cases throughout the day based on the exact supply and demand at a given time and governed by algorithms and software.
2) Dynamic pricing for competing commercial users will bring the industry closer to a free market with price discovery, but some protection will remain as a public good.
3) Certainty will replace uncertainty within the last mile. All drivers, whether citizens or commercial, will know the rules, rates, and availability before they get there.