Key takeaways
See our roundup of condition monitoring software that catches drift early.
Every asset arrives with specifications: maximum load, recommended speed range, ambient temperature limits, duty cycle, lubrication interval. These specifications collectively define an operating envelope, a region in the multi-dimensional space of operating conditions where the manufacturer says the asset will perform as designed and degrade at the rate the maintenance schedule was built against.
The envelope is usually documented in the manual, sometimes printed on a plate on the asset, occasionally on a wall in the maintenance office. It is almost never visible in the production data the operators see during a shift. The asset's actual operating conditions are recorded by the OEE system and the sensors; whether those conditions are inside or outside the envelope is rarely calculated.
The result is that an asset can spend hundreds of hours per month above its designed load, and nobody knows. The PM schedule, built for designed load, fails to keep up with actual wear. Failures look random; they are actually the predictable consequence of out-of-envelope operation. The article on manufacturing KPIs covers the related component trends.
The asset is operating within the conditions the manufacturer designed for. No special tracking. The PM schedule is calibrated for this region and should keep up with the wear rate.
The asset is operating beyond the conservative design envelope but inside a wider operational tolerance where failure is more likely but not imminent. Every yellow-tier event is logged. A weekly review identifies which assets are spending the most time in yellow.
This tier is where most plants discover their hidden wear. An asset that operates at 95th-percentile load 30% of the time will fail sooner than the PM schedule expects, and the yellow-tier tracking is what surfaces the pattern.
The asset is in a region where failure or damage becomes likely if operation continues. The system triggers an immediate operator alert, a maintenance work order, and depending on the asset class, either a forced stop or a documented decision to continue with explicit acceptance of the risk.
Red-tier events should be rare. If they are happening more than a few times a month on the same asset, the operational envelope is wrong for the actual production conditions, and either the production needs to change or the envelope needs to be renegotiated with the OEM. The article on work order management systems covers how red-tier events generate work orders.
Three sources, in priority order:
The starting point. Most manuals give a "recommended operating range" (green), a "maximum continuous" (boundary between green and yellow), and a "maximum intermittent" (boundary between yellow and red). Use these as the default if no other information is available.
For asset classes the plant has been running for years, the failure-history data often shows where the actual yellow boundary should be. A bearing class that consistently fails well below its rated maximum load is telling the plant the OEM's "maximum continuous" is too generous; tighten the green/yellow boundary accordingly.
Some products require operating in what would normally be a yellow tier. If a large share of the plant's production is in this category, the envelope itself may be wrong rather than the operation. The decision then becomes operational: derate the asset, run reduced PM intervals, or upgrade the asset. The piece on root cause analysis covers how to investigate envelope-process mismatches.
The asset has been running above its rated load for so long that everyone considers it normal. The PM schedule is calibrated against the rated load; the actual wear is calibrated against the operating load. The gap accumulates as a slow rise in failure rate that nobody attributes to envelope violation.
A specific SKU pushes the asset into yellow every time it runs. Operators do not flag it because it is a documented operating condition for that SKU. The OEE looks fine. The wear from that SKU disproportionately drives failures, and the failures appear randomly distributed across product runs unless the analysis specifically segments by SKU.
To hit shift production targets, operators run the asset faster in the last hour of the shift. The behaviour is consistent across shifts. The cumulative time in yellow is small per shift but compounds across the year into significant accelerated wear. The piece on the preventive maintenance schedule covers how this pattern shows up in PM cycle data.
The most insidious. The asset's actual operating conditions diverged from the envelope years ago. The operators who set up the current conditions are gone. Nobody on the current team knows what the original envelope was. The asset is operating in chronic yellow that nobody recognises because the institutional memory is missing.
The envelope-tracking system needs three things:
For a plant with 50 critical asset classes, this is roughly 50 envelope rules and 50 alert configurations, achievable in 6-10 weeks of focused work, sustainable thereafter with quarterly review.
The envelope-tracking concept works in any platform that streams operating-condition data and supports threshold rules. Where a unified OEE + CMMS platform helps is that the yellow-tier logging feeds the same failure-analysis pipeline as the work orders, so the correlation between time-in-yellow and failure rate is a query rather than a quarterly investigation.
Fabrico is built so envelope tier is a live status on the asset, visible to operators and to the maintenance team. To see what your envelope-violation picture would look like, book a demo .
Start with the plant's historical operating range, the conditions the asset has been running at without obvious failure for at least 12 months, as the green tier. Tighten as failure-rate data accumulates. The first envelope is approximate; the third is well-calibrated.
The disagreement is itself data. If operators consistently push into yellow because the production target requires it, the envelope is in tension with the production rate. That is a structural issue to surface, not a discipline issue with operators.
Envelopes apply per process step rather than per shift. The asset's envelope during charge is different from its envelope during run. The tracking infrastructure is the same; the rule definitions are more granular.
Yes. A live indicator at the operator station that shows green/yellow/red moves behavior faster than any retrospective report. Plants that surface the tier visibly find that operators self-correct within a shift; plants that only review tier weekly see the same patterns repeat.
Setting yellow boundaries that are so tight the asset spends most of its time in yellow. The tier loses meaning, the operators learn to ignore the alert, and the system stops producing useful data. The yellow boundary should be set so that an asset operating normally is in green at least 85% of the time.
Login
Blog
