This MIT Startup, AirWorks, Aims To Be The Top Aerial Data Analytics Service For Construction Firms
If you want the best-in-class startup advice delivered to your inbox, please fill out my survey for my upcoming mailing list here: Founder to Founder Newsletter Survey. The mailing list will launch in early October.
In the Age of Big Data, humans generate 2.5 quintillion bytes of data each day. Looking at data related to land surveying for construction, the lack of automated information processing makes the activity expensive and tedious, costing clients an average range of $10,000 to $20,000 for a two-month project. David Morczinek and Adam Kernowski established AirWorks, a Cambridge, M.A.-based data analytics startup that automates the processing of 20 million points of data from drone images to help decision-makers at land development and construction make sense of their aerial data.
Drones aren’t just for recreational use. They are playing a more significant role in a variety of commercial applications such as aerial filming and package delivery. The global management consulting firm, McKinsey & Company, estimates that the commercial drone industry will have an economic impact ranging from $31 to $46 billion on U.S. GDP. Morczinek, 31, and Kernsowski, 37, founded AirWorks to not only speed up the aerial mapping process via drones, but also automate the post-processing of the data to produce accurate computer-aided design (CAD) models that civil engineers and architects can immediately use. AirWorks is currently a participant in MIT’s premier startup accelerator, Delta V. Delta V has produced notable companies such as with LiquiGlide, Nima, Accion Systems, and Infinite Analytics. LiquiGlide, the platform startup responsible for creating permanent, slippery surfaces, has partnered with Elmer’s Products, and also raised $16 million from venture capital firms including Structure Capital, Valia Investments, Struck Capital and Pilot Grove.
Morczinek and Kersnowski incorporated AirWorks in August 2017. Kersnowski’s background in construction and piloting drones gave him confidence that drones were useful and viable in inspecting buildings and surveying construction sites. Morczinek, using his aerospace and data analytics background, intuited that drones could produce large volumes of data from the images they captured. That data could then be processed and presented in an immediately accessible format to construction companies and engineers.
The two cofounders did their due diligence by interviewing more than sixty construction and engineering firms. Morczinek, Kersnowski, and their team found that for a given hundred-acre plot of land, it would take two professional land surveyors approximately six weeks to survey the area. Also, current firms do not use drones as they lack the know-how to turn the images the drone collects into formatted data that they can immediately use.
“We knew that there was a gap between what was possible versus what goes on in the industry, and we’ve positioned AirWorks to be the bridge between the two sides,” says Kersnowski.
AirWorks uses retail drones installed with cameras designed take wide, high-definition aerial shots. These drones fly over a parcel of land to collect images of the site. These images are then stitched together to produce a point cloud. This point cloud contains various levels of data, which is then processed and exported into user-friendly formats for the startup’s customers.
Morczinek, the aerospace and data analytics half of the duo, is responsible for the post-processing of the data. His other half, Kersnowski, uses his construction background and piloting experience to operate the drones to capture the images they need efficiently. The cofounders and their team have built a proprietary machine learning algorithm to classify points, determining whether a given point is a bridge, building, road or other important feature to identify. The algorithm, coupled with computer vision from the drone, can then connect the points to resemble the original landscape.
The reason why Morczinek opted for a machine learning-driven approach is that the data that is collected from the images captured by these drones is unstructured, i.e., it is obtained in an unorganized format. Traditional classification algorithms are designed to work with structured data, rendering them unsuitable to meet AirWorks’ needs. Morczinek realizes the limitations of machine learning as well, such as classifying the wrong points based on poor learning from training data, so he and his team use their current combination of traditional and machine learning algorithms to process 20 million points of drone image data rapidly.
“Our goal is to make aerial data useful for decision-makers,” says Morczinek.
Currently, AirWorks can reduce the land surveying time from months to days. With a set of drones, the startup can map 400 acres in one day. After mapping, they can reduce the time it takes to process the data from two weeks using a team of trained engineers to twenty-four hours with only their proprietary code. This critical capability makes them attractive to construction companies throughout New England that could use their services. AirWorks’ ability to perform recurring inspections with drones every week versus once or twice more via regular methods dramatically reduces the costs of rework. Rework is roughly 5–10% of the cost of developing on new land. By eliminating rework, the cost savings alone are roughly $75 billion per year. AirWorks’ primary source of revenue is in the cost and time reduction of errors from land surveying and mistakes during construction that led to rework.
AirWorks can do more than just producing maps. After the construction of a building is finished, the drones can be used for aerial inspection of the building, which is critical for quantifying the risks the property faces for insurance purposes. The startup isn’t the only one eying the construction space for drone usage. Some of AirWorks’ competitors are Drone Deploy, Kespry and 3DR. The Cambridge-based startup’s main strategic advantage is that they deliver custom-formatted drone image data that their clients can use immediately, instead of just producing models and images like their competitors.
In the future, AirWorks hopes to be the go-to analytics provider for the construction and surveying industries. Their goal is to automate most, if not all, of the aerial analytics produced by drones. Right now, their current service offerings are best suited for a construction firm’s early phases of design and project planning. Also, the team hopes in the long term that they can provide regular site and development monitoring solutions for their clients.
With startups such as AirWorks tackling the challenges of Big Data through aerial means, we collectively get one step closer to understanding our physical impact on the world through the digital footprints we leave behind.
If you enjoyed this article, feel free to check out my other work on LinkedIn and my personal website, frederickdaso.com. Follow me on Twitter @fredsoda, on Medium @fredsoda, and on Instagram @fred_soda.
This work was first published on Forbes.