“Hands-Free” Fully Autonomous, Plant-Scale, Anomaly Detection AI

This paper presents an all-new self-supervised autonomous “plant-scale” artificial intelligence (AI) - capable of monitoring every parameter and asset of a plant - providing automated detection and diagnosis of anomalies in steelmaking. Automatic anomaly detection informs plant operations of conditions that otherwise would be undetected, leading to informed production and maintenance decision-making. Self-supervised AI overcomes the challenge machine learning faces scaling to the needs of steel manufacturing by accommodating the challenges of constant equipment, environment, and product changes that thwart classical supervised machine learning. This paper will show this new AI in commercial operation today.

Joe Porter | Falkonry Inc.
 

Keval Bhanushali | Falkonry Inc.
 

Dan Kearns | Falkonry Inc.
 

Nikunj Mehta | Falkonry Inc.