NOVEMBER 13, 2017

Steering Predictive Maintenance with Machine Learning


Join Petasense at IMC-2017 to learn about the merits of machine learning for predictive maintenance.

With self-driving cars getting ready to hit the roads, the lines between science fiction and reality are blurring. “Knight Rider” is coming to life with practical applications in the real world. Artificial intelligence (AI) and machine learning are driving this technological advancement.

Petasense is at the forefront of automating predictive maintenance in the industrial world, using the same technology. Just like how machine learning predicts the behavior of a pedestrian and instructs the autonomous driving vehicle, it can also be used to predict the behavior of a machine and provide actionable intelligence to reliability and maintenance folks. How cool is that?

“Machine Learning” is a class of algorithmic techniques that uses data to imbue computers with the ability to learn without being explicitly programmed. When properly trained, these techniques build generalizable models that can make accurate predictions, find unknown patterns, and offer deep insights.

Simon Xu, a data scientist from Petasense will lead a talk on “Practical Machine Learning for Predictive Maintenance” at IMC-2017 on Monday, Dec 11. The session will demystify how machine learning applies to predictive maintenance and the steps required to create a successful machine learning based reliability program. The talk will also discuss how to best evaluate a machine learning program in terms of how it contributes to the bottom line by presenting use cases where predictive analytics saved industrial customers over a million dollars.

Make sure to stop by to learn about:

    • The art of “feature engineering”
    • How to scale and adjust data in a way that gives you the most useful results
    • The economics of maximizing your ROI by preventing unplanned downtime
    • Machine health scores and the best anomaly detection algorithms
    • Making smarter maintenance decisions with machine learning

So mark your calendars. As reliability leaders and innovators, you definitely don’t want to miss this.