Published: September 29th, 2017

Toronto – At the 2017 CDN Channel Elite Awards the Vancouver-based Sierra Systems became the first ever Disruptor of the Year award winner.

VMware Canada Channel Chief Tara Fine said traditional industries are partnering with IT companies such as Sierra Systems to leverage the IoT with predictive analytics, often from existing data. Developed properly, the newly available paradigms for business analytics, such as machine learning, can lead to new processes, and massive cost savings. Fine presented the award to Paul Twigg, vice president of technology, and Anthony Bulk, principal, of Sierra Systems.

“It’s an honour for us to accept the award, to be a technology disruptor in Canada,” Twigg told CDN after accepting the award. “We believe in innovation and we are trying to bring that innovation to our clients every day. From machine learning, predictive analytics, AI, IoT, and all the different technologies, we try and help them built to leverage those technologies and propel their businesses forward.”

Sierra Systems disruptive solution supports energy and utility companies by searching pipelines for leaks and damage. This was done by implementing the SmartBall solution through a pipeline, and recording audio data with a high-fidelity microphone. These sound recordings were then copied to a lab and interpreted by technicians to identify potential leaks, and assessed for severity. Sierra Systems used a deep learning artificial intelligence system to vastly improve what was a tedious manual process. The SmartBall data is now fed into a model developed using a commercial grade cognitive toolkit for deep learning.

The hardware aspect of the SmartBall solution comes via a partnership with Pure Technologies who created the hardware component.

“This is really a great story about people and technology working together to solve a real world problem. We partnered with Pure Technologies, a world class leader in developing innovative solutions, to detect and monitor infrastructure,” explained Bulk.

The machine learning model uses the latest in visual classification techniques and Convolutional Neural Networks to distinguish leaks from non-leaks, removing the manual process, maintaining the same accuracy, and completing the task 150 times faster than a person could.

The model is compact enough to run on devices out in the field where the pipelines reside. With this data, managers can execute repair orders in hours instead of weeks.

Sierra Systems also won Gold for Best Internet of Things Solution.