Mtell: Preventing Failure, Mitigating Risk and Driving Performance

It was in the middle of his career at one of the largest manufacturing companies, when Paul Rahilly came across the significant pain points that existed with critical equipments in terms of equipment liability, breakdown, operational expense, maintenance strategy, and integrity issues. Later on, while turning his focus to help organizations eliminate these pain points and switch to a much greater visibility in terms of operation and connection to the real time data source, Paul founded Mtell. Established in 2006, Mtell is driven by a consistent and costly pharmaceutical industrial issue— equipment breakdown.

Headquartered in San Diego, CA, Mtell is at the forefront of Machine Learning for industrial maintenance. The company’s flagship product, Mtell Advanced connects machines with operations and maintenance systems and engineers and uses Machine Learning with its Predictive Analytics on large manufacturing equipment. The mature and refined technology delivers a smooth, hassle free and time saving function by analyzing millions of sensor data point to rapidly converge on behavioral patterns. It classifies anomalies as new normal states or previously undetected faults and detects patterns as minuscule interaction between multiple sensor streams. Mtell Advanced also analyzes machines to give alerts to maintenance systems for servicing to avoid equipment break down. “Mtell’s focus to bridge the gap between operations and maintenance resulted in the inception of Mtell Basic. Later on, while innovating our technological footprint, we launched Mtell Advanced with Machine Learningbased condition monitoring,” says Mike Brooks, President, COO.

Mtell differentiates itself through its technological aspect and its sound nderstanding of the ecosystem that gives it the ability to accommodate to changing scenarios. The company has a specific close loop system that looks at each individual machine and improves its operation over time with less human effort. Mtell doesn’t use any models, as it captures behavior by itself using neural network Machine Learning analysis. It also learns and adapts to changes in operating conditions. “When it comes to the machine, we look for what’s normal and what isn’t, using pattern recognition. In the event an organization faces equipment failure, Mtell can give as many as 30 days warnings or more of impending failures, where the notifications are very accurate and precise,” says Brooks. Mtell has pioneered in the reliability space of providing Machine Learning solutions.