A reject rate climbs, an energy figure drifts, a cycle time creeps. Munshi surfaces the deviation against what the asset normally does, and shows the signals that moved alongside it on the same clock — the current that rose, the delay that was logged. It shows that they moved together. Your team confirms the cause; Munshi does not declare one.
Your ERP shows the outcome drifted.
It can't show you what moved with it - the signal, the delay, the condition that shifted on the same shift.
Why the gap existsWhat it does
An outcome is judged against what this asset normally does, not a plant-wide target — so a drift that matters on this line is not lost in an average.
When an outcome moves, Munshi surfaces the signals and logged reasons that moved alongside it, time-aligned — the material for a diagnosis, laid side by side.
Munshi shows that two things moved together. It does not tell you one caused the other. The operator confirms the cause, and the confirmed cause is written back to the asset’s record.
In the plant
Energy per tonne deviated; the furnace held and a crane wait was logged on the same clock. Munshi shows both moved together and lets the operator confirm the cause.
A reject rate climbed as a die approached its life. The signal and the outcome sit side by side; the person decides.
A repair rate rose on one shift alongside a drop in arc-on time. Surfaced together, for a human to weigh — not ranked as a cause.
What Munshi surfaces
Teal is normal, maroon is flagged. Illustrative data.
The flagged bars are the deviations, not the average.
The flagged event is what the operator logged — shown alongside, not blamed.
Surfaced for a person to weigh, not ranked as a cause.
The pattern a single daily number can’t show.
Get started
See this on your own floor.
Bring a month you already understand. The useful test is whether Munshi finds what you already know — and then what you didn’t.