Steelmaking is a complicated and extensive process with hundreds of equipment working in tandem and producing terabytes of data every week. Unexpected equipment failures can result in production losses in the range of millions of dollars. Such events can be prevented if the operations team has plant-wide operational visibility combined with fact-based insights that allow smarter decisions.
In this presentation, we will discuss the real-world applications of time series AI across the steelmaking processes such as casting, hot and cold rolling etc. One such challenge is strip-break during cold rolling of steel that results in yield loss due to line stoppage, re-work, and may also cause damage to equipment. The presentation showcases how time-series AI automatically classifies strip break events and provides underlying root causes. This enables the operations team to implement corrective measures quickly and prevent repeat occurrences of strip breakage, thereby improving production efficiency.