Making sense of millions of data points for manufacturers

The why, not just the what

The why, not just the what

We built a system for operators to classify spans of time and data in realtime, providing Ops Teams and Process Engineers context for decision-making, and allowing for correlation of raw machine process (“speed = 907RPM”) metrics to events (“mechanical failure”).

This provides operations a complete picture of operational health, and ultimately enabled more robust monitoring, reporting, analysis, and machine learning applications.

Since its launch, we’ve observed customers reduce their “unknown time“ from 100% to low single-digits in a manner of months, and have been informed of million-dollar defects that were avoided through early visibility afforded through this system.

factory.png
”What if …

”What if …

… operators had the ability to add the ground-truth context of the operating state, complementing the raw information from the machine?”

 Wireflow sketches of the experience, exploring automatic event prompting  conflict resolution, and offline connectivity behavior.

Wireflow sketches of the experience, exploring automatic event prompting conflict resolution, and offline connectivity behavior.

operator-wireflow-b.png
operator-auto-transition.png
 The shipped product, showing the correlation of machine process data (the  what ) with operator-input state information—the context, the  why .  This allows operations to gain an immediate understanding of what’s going on in their plant, quickly triaging interesting or problematic events, as well as retaining a permanent record of the.

The shipped product, showing the correlation of machine process data (the what) with operator-input state information—the context, the why.

This allows operations to gain an immediate understanding of what’s going on in their plant, quickly triaging interesting or problematic events, as well as retaining a permanent record of the.