In this Section
11-07-2016

# CertainError: Spotlight on Startups at Clarkson University Shipley Center for Innovation

When Clarkson University Associate Professor of Mechanical & Aeronautical Engineering Ronald S. LaFleur realized his students weren't fully grasping critical topics such as error propagation and uncertainty analysis, he devised a more effective way of teaching. When it came to sharing that knowledge with a wide range of industries, he turned to Clarkson's Shipley Center for Innovation for help.

The result of that collaboration -- CertainError -- is just hitting the market. Here's the thing to remember. His company makes products that are useful to everyone, not just rocket scientists or mathematicians or students.

The company website, CertainError.com, has a calculator help menu and tutorial videos, as well as posts of recent projects that serve industries ranging from engineering, finance, and math, to medicine, science, sports and statistics.

To reach this new way of thinking, LaFleur had to embrace the reality of error, meaning that all the numbers we deal with in life aren't exact because we lack an accurate way of measuring them. Temperatures, speeds, return rates, projected costs and dimensions all provide just ballpark generalizations. The professor aims to help define the parameters of that ballpark.

“Consider how we measure temperature,” he says. “We look out the window on a cold day and the liquid is visible inside the glass of the thermometer. By looking, you're measuring temperature in a mini experiment. Suppose it says it's 32 degrees Fahrenheit. That's the normal approach. I'm saying, attach the error to this figure. The fluid doesn't know there is a number on the scale, it's just reacting to air temperature and rising or falling. By us rounding to 32 degrees, the true temperature is not exactly 32 degrees.” For this example, knowing air temperature near the freezing point can be very important.

“The actual temperature is 32 degrees Fahrenheit, plus an amount of error,” he says. “The truth is better expressed with the temperature reading along with the maximum error to indicate there is a limit to that knowledge.” To help the public better interpret numbers in business and everyday life, his company's products include the CertainError Calculator app for Apple and Android devices, and CertainError software, in three formats.

Non-scientists and math-phoebes may be quick to say: “So what? Cold is cold.” Ask yourself, though, how accurate would you like medical tests and lab results to be? How certain do you want to feel about an airplane's stress testing? How much do you need to trust a weather report when a hurricane is approaching? How accurate is that political poll the news is reporting? How much confidence do you have in your fantasy football picks or search engine results? And so on.

Or, let's go back to the students. Lab courses that LaFleur and his colleagues routinely teach include experimental methods. In one simple experiment, students will drop a ball to demonstrate the acceleration of gravity.

“Suppose instead of just dropping the ball, they also learn how to release a ball and measure its drop time accurately. They take away a general tool, such as ‘measuring time accurately.’ When they graduate and go to work, now they'll have a tool that has wider value than just the determination of acceleration of gravity. They have to learn how to measure and evaluate data,” he says. “I want them to be aware of errors and learn how to control errors.”

How LaFleur came to this frame of mind leads to the reason he started CertainError. While he was pursuing his doctoral degree at the University of Connecticut, jet engine manufacturer Pratt & Whitney was one of his sponsors. Then in 1988, he came to Clarkson. Around 1997, he went on sabbatical and chose to work for a year with Pratt & Whitney in Florida.

The result was six patents, he notes, but this experience also gave him additional insight into the different attitudes between educators and businesspeople.

“With professors, knowledge is our product; we're generating new knowledge and communicate knowledge. For patents, it's the opposite. You have to protect the idea before you can share it. When I was at Pratt & Whitney, information was very controlled and serving the customer was at the forefront,” he says. “When I got back to campus, I had a different perspective about what professors do.”

Looking at the struggles students were having, he came up with an approach to error propagation and uncertainty analysis that uses geometry and algebra instead of differential calculus and statistics. To work on this issue, he had to read a lot of math, teach himself and develop what was missing. He also had to embrace personal uncertainty, with the risk that this new approach would not work. He worked alone at first and then began testing his ideas in class.

Around 2012, he began talking to colleagues about it. He knew his methods and now needed to communicate it. People suggested he submit a provisional patent and start a business through the Shipley Center for Innovation.

“This is not just a teaching tool,” LaFleur emphasizes. “It could change the way people use numbers and calculate. It really has wide implications so I wanted to start a company to commercialize it and share it. I disclosed my inventions to Clarkson University, asking to partner with them.”

“What I have today is due to the Shipley Center and the people they contract to,” he affirms.
“This isn't just a place to start a business, it's an incubator. They help you grow. I needed to think like a businessman. They got someone to help with a logo and the website and the app for phones. I can do the math and write software but I didn't know how to make an app and market it.”

LaFleur's CertainError website went up in September and the apps followed two weeks later. He says he has no doubt that partnering with the Shipley Center was his best choice.

Matthew Draper is executive director of Clarkson University's Shipley Center for Innovation, which guides innovators like LaFleur through the complex process of commercialization, providing them with the necessary tools to transform ideas into reality. The types of assistance include mentorship, networking with Clarkson alumni, marketing, branding, legal counsel about intellectual property rights and processing for patents, copyrights and trademarks.

About 30 new startup companies are created through the Center every year, and another 75 to 100 projects could be startups, Draper has noted. More are always welcome.

To understand how error and uncertainty affect the world we live in, explore the website and tutorials at www.certainerror.com.