How to detect gradual efficiency decline

Hi ​@TIM,

For slow, multi-month drift like this, the best home is an overview dashboard for the cooker: turn each symptom into a per-cycle KPI, trend it over a long timeframe so the slope is obvious, and put a monitor tile on the same board for the early warning. Fingerprints aren't the tool here they catch a single odd cycle, not a creeping trend across many.

What to put on it (one value per cycle, from your cycle-identification search):

• Temperature rise slowing → time-to-setpoint, or the heating-stage duration.
• Steam usage rising → steam consumed per cycle (integral of steam flow), normalised per unit product.
• Cooking duration → the cycle's own duration.

Trend each over months — one point per cycle, so you sidestep the 24-hour aggregation cap and just read the drift.

DashHub tips & tricks:

1. Use the Trend (shared-axis) view, not Stacked, for the KPI lines so they overlay and you can compare slopes stacked turns them into unreadable micro-lanes.
2. Set the Global Timeframe to months, but know it only drives TrendHub and ContextHub view tiles current-value, gauge and monitor tiles keep their own scope. That's the #1 "why won't my dashboard sync" surprise.
3. Add a status indicator: a current-value tile with conditional colouring (green→red), or the Gauge tile on 2025.R3+, fed by your health KPI so degradation reads as a colour, not just a line.
4. Add a Monitor tile for the early-warning monitor, so detection and alerting sit on one screen.
5. Use a Text tile for the thresholds/playbook ("review cleaning when steam-per-cycle > X") and an External tile for the P&ID or an equipment photo for context.
6. Build it as an engineer/manager rollup and share it through Presentation mode so the team sees a clean, read-only board.

The DashHub – Create a Control Center of your Process recording webinar walks the full build. For hands-on practice, work through the basic guide DashHub exercises and the intermediate guide DashHub exercises.

Kind regards

Frederik