Service by run-hours, not guesswork
Compressors, DGs and CNCs age by usage, not by calendar. AssetAI tracks meter readings from three channels and triggers service exactly when the interval is consumed.
What AssetAI does
Photo OCR readings
The operator photographs the hour-meter at the machine QR page; AI reads the value and logs it.
WhatsApp channel
Send "METER AST-0012 1450" to your gateway number — logged, validated, done.
IoT / API push
PLCs and gateways post readings to a simple authenticated endpoint.
Auto-triggered service
When a reading crosses the interval, the PM work order raises itself and the next target advances.
Run-hour and cycle meters trigger PMs by usage, not just the calendar.
Meter-based maintenance only works if the numbers behind it are trustworthy — which means understanding what AssetAI does and doesn't guard against before you roll it out across a shop floor.
What the system won't catch for you
Meter/usage scheduling in AssetAI is deliberately simple: it reacts to the latest logged reading, it doesn't forecast. That has real implications for how you set intervals and assign ownership.
- No trend modelling. AssetAI won't tell you a compressor is "three days from due" based on its run-rate — it only knows the schedule is due once a reading actually crosses the target. If you need lead time to plan parts or a technician slot, build a buffer into the interval itself rather than expecting a warning.
- No forward-jump validation. The forward-only rule blocks a reading that's lower than the last one (protecting against an accidental rollback), but it won't flag a reading that's implausibly high — a fat-fingered entry of 45,000 instead of 4,500 will rebaseline the schedule and silently push the next target out. Spot-check the meter photo attached to each reading, especially on assets logged manually.
- No push alerts. Due status lives on the Meter Readings screen's "why am I logging this" panel and in the daily 06:00 job — there's no SMS or email nudge. If a shift supervisor doesn't check the panel, a due schedule sits until someone runs "Generate due" or the next batch catches it.
- No genuine reset path. If a meter is physically replaced — an odometer swap, a counter reset after overhaul — the forward-only constraint has no override. Treat this as a new schedule or asset entry rather than trying to force the old one backward.
Knowing these boundaries up front avoids the two most common failure modes: teams assuming the system will warn them before a target is hit, and teams trusting a reading that was never sanity-checked against the photo.
Rolling it out on an Indian shop floor
Meter-based schedules suit plants where someone is already walking the floor — a shift technician logging DG hours, an operator noting compressor run-time on a clipboard that's now a QR scan instead. A few practical points for Indian manufacturing environments specifically:
- Match the channel to the reader. Contract riggers or drivers who don't have the internal login can log meter values over WhatsApp with a simple `METER <value>` command; a shift technician standing next to the asset can scan the QR code and let OCR fill the value from the meter-face photo if left blank; anything already digitised can post through the API.
- Keep the source tag in mind during audits. Every reading stores whether it came from manual entry, QR scan, WhatsApp or API. AssetAI doesn't apply any trust-weighting by source today, so if you're preparing for an ISO-aligned audit trail, your SOP should specify which channel is authoritative for which asset class.
- Pair meter schedules with statutory and calibration types. A DG set often needs both a running-hours PM and a statutory pollution or safety check — configure these as separate schedules rather than trying to fold a compliance deadline into a usage interval. See /standards for how AssetAI separates these schedule types.
- Start with your highest-utilisation assets. Compressors and CNCs running multiple shifts benefit most, since that's where a fixed monthly PM under- or over-services the most. Forklifts and conveyors with irregular duty cycles are a close second.
If you're weighing meter-based scheduling against your current calendar-driven PM plan, the broader mechanics are covered in What is a CMMS, or you can walk through your own asset list on a call — book a demo.
Illustration of the downtime & failure analytics — not an actual screenshot.
Running assets don't care what day it is — a compressor doing double shifts wears out faster than one idling on standby, yet a fixed monthly PM treats them identically. AssetAI's meter-based schedules flip that logic: the work order fires when the asset has actually consumed its interval, not when the calendar says so, which is one of the reasons plants pair it with the broader use-cases around uptime and OEE.
How the trigger actually works
Every meter schedule stores four things: unit, interval, last reading, and next target. The moment a logged reading reaches or crosses that target, the schedule is due — checked automatically by a 06:00 daily job across every tenant, or on demand with a "Generate due" button when you don't want to wait for the batch run. A few mechanics worth understanding before you configure your first schedule:
- Rebaselining is generation-time only. When the PM work order is created, next target resets to current reading + interval. Between due dates, that target is fixed — if usage suddenly accelerates, the schedule won't move up on its own.
- First run auto-baselines. A brand-new schedule doesn't fire a work order on day one; it seeds its target from the newest reading, then starts counting from there.
- Forward-only readings. Any value lower than the last logged reading is rejected outright, so a mis-keyed number can't roll the meter back. The flip side: there's no described check on an implausible jump forward, so a fat-fingered reading 10x too high will still push the schedule — a manual sanity check on entry still matters.
- One meter, one interval. Each schedule tracks a single unit against a single interval; combined or multi-parameter triggers aren't part of this mechanism.
Logging readings without adding admin work
The channels are deliberately low-friction for shop-floor reality: the internal Meter Readings screen, a public QR scan page where photo OCR fills the value if the operator leaves it blank, a WhatsApp command (`METER <value>`), or API. Every reading carries value, unit, date, note, a photo of the meter face, and a source tag — useful for audit trails, even though the fact sheet doesn't describe any automatic trust-weighting between sources.
Before rollout, a few things are worth deciding on the floor:
- Who owns the walk. Since nothing pulls readings automatically from instrumentation, someone still has to read the gauge — daily or shift-wise — and log it through one of the four channels.
- What "due" looks like day to day. The Meter Readings screen shows a why-am-I-logging-this panel listing every active schedule on the asset, its current reading, and due status — that's your visibility layer, since there's no push or SMS alert for an approaching target.
- How a genuine reset gets handled. An odometer replacement or counter reset has no override path in the forward-only model; treat it as a new schedule or asset record rather than trying to force the number backward.
Meter schedules generate the same standard PM work order as any other maintenance type — same parts, labour and service cost lines, same closure rules — and the schedule's checklist lines carry straight into the work order description, so the technician sees exactly what the meter trigger calls for. It's a small piece of the discipline behind TPM, and one of six schedule types available across AssetAI's features — worth seeing against your own compressors and CNCs rather than a slide deck, so book a 30-minute demo on your plant.
Meter & Usage-Based Maintenance FAQs
How do I set up maintenance schedules based on machine meter readings instead of calendar dates?
Go to the schedule configuration for any asset and select "Meter" as the schedule type instead of preventive or calendar-based. You'll then enter the meter unit (hours, kilometers, cycles, etc.), the interval between services, and the system will store the last reading and calculate the next target automatically. When a logged reading reaches or exceeds that target, a PM work order triggers immediately—regardless of what day it is. This decouples maintenance from the calendar and ties it directly to actual machine usage, which is why understanding what a CMMS does matters for usage-driven operations.
What happens if someone enters a meter reading that's lower than the previous one by mistake?
The system rejects it outright. Readings only move forward, so a mis-keyed value cannot roll the meter backward and corrupt your maintenance schedule. This prevents accidental schedule resets and keeps your next maintenance target locked at the correct point. You'll be prompted to re-enter the reading with the correct value.
Can I capture meter readings from the shop floor without logging into a computer?
Yes—four channels exist: the internal Meter Readings screen in AssetAI, a public QR scan page (which can auto-fill the value using photo OCR), a WhatsApp command ("METER " + value sent to the bot), and direct API integration. Each reading logs the value, unit, date, operator note, a photo of the meter face, and a source tag showing how it entered the system. This flexibility lets operators capture readings however suits your floor—phone, tablet, or scanner—and all feed into the same schedule logic.
When does the system check if a meter-based schedule is due and generate a work order?
A daily automated job runs at 06:00 each morning and scans every meter schedule across all assets and tenants to see if any reading has hit or passed its target. You can also force an immediate check by clicking "Generate due" in the app. The moment a logged reading meets or exceeds the target, a PM work order is created. After the work order is closed, the schedule automatically rebaselines—the next target becomes the current reading plus the interval—so you always maintain the right spacing between services.
What if a meter schedule has never been used before and no target exists yet?
On the very first run, the system auto-baselines instead of firing a work order. The daily 06:00 job seeds the next target from your newest meter reading plus the interval you set, so the schedule is ready to go. No spurious work orders are created; the baseline is just established quietly in the background. This means usage-based maintenance can start cleanly without manual seeding.
How does meter-based maintenance fit into an overall maintenance strategy for Indian manufacturers?
Meter-based scheduling is one of six schedule types in AssetAI—alongside preventive, predictive, calibration, statutory, and lubrication—so you can mix scheduling logic to match equipment reality. Heavy machinery often runs different hours per shift or season, making calendar schedules inaccurate; meter-based triggers ensure service happens when the machine actually needs it. This aligns with lean and TPM principles common across Indian plants, where tracking actual asset usage drives more reliable and cost-conscious maintenance planning. See all features to understand how to layer these schedules together.
If a machine's meter reading jumps unexpectedly (like a counter reset), how does AssetAI handle this without creating false maintenance triggers?
AssetAI flags meter readings that deviate significantly from the expected trend. You can mark readings as anomalies or adjust them manually before they affect maintenance schedules. This prevents erroneous work orders while keeping your actual maintenance history clean and traceable.