The Future of Street Maintenance is Here
Will we realize it in time for Biden’s big infrastructure investment?
TLDR: President Biden has released an ambitious, over $2 trillion infrastructure plan, and the United States will likely soon invest trillions of dollars in rebuilding its infrastructure. $115 billion in the plan goes to fixing our nation’s highways, bridges, and roads, including fixing the perennial problem of potholes. Currently 77% of roads across the country are managed by local governments and have very little to no data on the actual street condition. New low cost sensor technologies that can be found in any modern smartphone enable the country to improve how our country manages and maintains its roads.
To explore this opportunity, ARGO convened a small group of experts in the field for a friendly weekend Saturday morning conversation. Because nothing says the weekend like pothole data!
Participants included:
- Mike Bloomberg, Urban Technology Researcher at Cornell Tech
- Ben Schmidt, CEO of Roadbotics
- Sam Edelstein, Former Chief Data Officer of the City of Syracuse
- Varun Adibhatla, Co-Founder of ARGO
Infrastructure week! For real this time!
It’s no secret that America’s infrastructure is crumbling. John Oliver had a great bit a few years back on Infrastructure week. He humorously pointed out that the American Society of Civil Engineers to assess the state of our infrastructure is “like having the state of our nation’s tennis balls assessed by the American Society of Golden Retrievers.”
First, this highlights the need for an independent objective measurement of the condition of our nation’s infrastructure. Second, despite their self-interest, the engineers are indeed correct and our country lags international rankings in infrastructure quality.
Lessons of California’s Big Investment in Road Repair in 2017
In 2017, California passed a new gas tax championed by then Governor Brown to repair the state’s roads. This investment was supported by the Save Our Roads coalition which produced maps of the state’s road conditions to highlight the importance of investment.
Source: Save Our Streets 2018 Report https://www.savecaliforniastreets.org/read-the-report/
The only problem? Those maps don’t necessarily actually reflect road conditions. The underlying measurements underlying those statistics are done by human inspectors conducting “windshield surveys” which is a fancy phrase for driving by and looking around.
That implicit bias in the measurements is increased by the fact that those human inspectors only survey a sample of the streets. That bias is further massively increased by the fact that not every city sampled in the SaveOurStreets Report actually conducted a street quality survey. Some just extrapolated from their neighboring cities.
It’s a classic case of garbage in, garbage out!
No amount of mathmagic can correct for the reality that the underlying measurements of street quality are biased by human error, often out of date, and incomplete. That’s not to mention the fact that these flawed data can cost millions of dollars in consultant expenses.
Luckily there exist numerous tools that provide a lower cost, higher quality result.
Low cost sensor technology
ARGO built an open source tool to use GPS, accelerometer and imagery data to measure street quality. Those pictures enable public works departments, mayors, and really any citizen to literally see the ground truth.
That technology was pioneered in New York, redeployed by ARGO, and thanks to the magic of open source and the internet redeployed by others as far afield as Italy and Santiago, Chile.
SQUID uses a vehicle mounted smartphone and open source technology to enable anyone to digitally survey the quality of streets and bike lanes. Learn more about how SQUID was developed here.
Numerous other tools exist such as the Germany based Vialytics. Google has also developed pothole image detection tools. Another promising such solution is the smartphone survey tool offered by RoadBotics, whose CEO Ben Schmidt participated in our conversation. Below you can see an example of their embeddable map feature, which can be shared to provide insights into any road network. This data is pulled from their interactive, pavement management platform, RoadWay.
RoadBotics recently produced a brief overview of road conditions in 20 cities across the US. Their 2021 US City Road Report provides an incredible apples-to-apples comparison of road conditions in downtowns across the nation. See below for an example image.
These approaches have long term benefits. Street maintenance proves the old adage that an ounce of prevention is worth a pound of cure. One often touted rule of thumb is that early maintenance in “good” or “fair” roads that need a simple resurfacing is 10x less costly as maintenance in severely degraded streets that need deeper restructuring.
This reality presents a tremendous opportunity for decision makers to meet the moment and not settle for the old status quo simply because that’s the “way things have always been done.” Below we present several recommendations to accelerate this transition to modern data collection technologies and better maintain city streets.
Recommendations for policymakers
- State and local governments looking to maximize the impact of future street maintenance investment should utilize a small portion of their one time fiscal recovery funding from the Covid Stimulus to map their street network using modern sensor technologies.
- The National Science Foundation should convene a working group including public works leaders, industry experts and academics to set a new national data standard for local street condition reporting that utilizes modern data collection technology. That standard should set conditions so that a road with a certain score has the same condition whether it was surveyed in Seattle, San Antonio, San Francisco or for that matter Sao Paulo.
- State and Federal infrastructure investment should incentivize cities to utilize modern sensor technologies moving forward. Future federal funding should be contingent with utilizing the data standard set out in recommendation two.
- Bond rating agencies and insurers should incentivize standardized infrastructure asset condition data as a factor in credit ratings.
