AMC & Contract Management

A technician raises a call on a stamping press, the vendor turns up, fixes it, and bills ₹18,000 — except the press was still under warranty for another four months. Nobody checked, because nobody had an easy way to check. Multiply that across a 40-machine unit with AMCs from six different vendors, and the leakage is real money paid for repairs that should have been free. AMC & Contract Management in AssetAI exists to stop that specific leak — by tying coverage to the asset itself and checking it automatically, every time a call is raised.

What lives on the asset

There's no separate "contract" record to maintain — AMC is stored as a set of fields directly on the asset: whether AMC applies, the vendor, start and end dates, the cost, and a schedule (monthly, quarterly, half-yearly, or yearly). Warranty dates and free-text notes sit alongside it. This keeps things simple for the common case in Indian plants — a compressor, a boiler, a CNC machine each carrying its own AMC — but it does mean if one vendor's contract covers 12 machines on your line, you'll enter that contract's dates and cost 12 times, once per asset. There's no master contract entity that rolls those up, so if your plant runs blanket service agreements across large asset groups rather than machine-wise AMCs, this module will feel like the wrong shape for your paperwork.

For most single-asset AMCs, though, it's exactly what a maintenance planner needs at the point of decision, not buried in a filing cabinet or a vendor's PDF.

Coverage is decided for you, not looked up

Every asset has a live coverage status, computed the same way every time, with a fixed order of precedence: in warranty beats under AMC, which beats expired, which beats no coverage recorded. You don't have to remember which comes first when both warranty and AMC dates happen to overlap — the system always resolves it the same way, so two different planners on two different shifts get the same answer.

When someone raises a service call against that asset, the call is stamped with that coverage automatically — under warranty, under AMC, or paid (chargeable) — and the vendor field is pre-filled from the asset's AMC vendor, falling back to its supplier if there's no AMC vendor set. That single stamp is what the rest of the workflow, and the approvals sitting on top of it, are built around.

The repair loop, from call to closed

Once coverage is known, the call moves through a fixed sequence with an automatic time trail: Requested, Visited, Quoted, Quote Approved, In Service, Completed, Billed, Bill Passed, Closed — with Cancelled available at any point. Every transition stamps who moved it and when, so if a vendor claims they were on site on the 14th, you have the actual timestamp, not a WhatsApp message to dig through.

• Paid calls are blocked from starting service until the quote is approved — no work begins on a chargeable repair without sign-off.
• Warranty and AMC calls skip that gate; they can move straight from Requested or Visited into service, because there's no chargeable risk to approve against.
• Only someone holding the service-call approve permission can approve a quote, pass a bill, or close a call — so a contract labourer logging calls on the shop floor can't also clear the spend.
• Entering a quote opens an approval request carrying the quote amount and routes it through whatever approval chain you've configured, rather than assuming one fixed approver for every plant.

This is deliberately close to how a work order already moves through your plant, but scoped to vendor jobs where an external bill is coming.

Catching the bill that shouldn't exist

This is the part that actually recovers money. If a bill above zero is entered on a call that's tagged as warranty-covered or AMC-covered, the system auto-tags the note: "[NOTE: asset was under AMC/WARRANTY — verify chargeability]". It doesn't block the bill or make the decision for you — vendors do sometimes have legitimate reasons to charge outside AMC scope, like consumables or damage from misuse — but it puts the flag exactly where the approver is looking, at the moment they'd otherwise just wave it through.

Once a bill is passed, the cost posts onto the linked work order and the original requester gets notified — so the person who reported the breakdown knows it's closed, and the cost is sitting where your maintenance-cost reporting expects to find it.

Notifications on raising a call, and on bill entry, go to everyone holding service-call approval rights, so there's no single point of failure if one approver is on leave or the plant is running a night shift with a skeleton crew.

Knowing coverage is about to lapse

A daily sweep checks every asset for warranty or AMC ending inside a window — 30 days by default — and raises a "Coverage expiring" notification with the message "Renew or plan for chargeable repairs." It won't fire on every check either; each asset alerts at most once every 6 days, so you get a nudge, not a flood. These alerts go to people holding breakdown approval, on the assumption that whoever signs off on breakdown spend is also the one who should be renewing an AMC before it lapses into paid repairs.

What this is not

Worth being direct about the edges: there's no contract document store or version history, no renewal workflow that generates a fresh AMC record when the old one ends, and no automatic generation of scheduled AMC visits — the monthly/quarterly/yearly schedule field is stored on the asset but nothing acts on it yet. There's no AMC invoicing or payment-schedule tracking, and while SLA hours can be captured on a call, nothing calculates a breach or a penalty against them. And there's no vendor-side rollup showing total AMC spend or contract value across your vendor base.

If your plant runs machine-wise AMCs and the real pain is repairs getting billed that shouldn't be, or vendor jobs going untracked between "call raised" and "bill paid," this closes that gap cleanly. If you need master contracts across many machines, SLA penalty math, or renewal automation, it's fairer to say so now than after you've loaded in your first hundred assets — worth checking against the fuller features list or the use cases that map to how you run maintenance today, or just book a demo and bring your ugliest AMC spreadsheet.

Getting the AMC data in before go-live

The honest cost of setting this up isn't configuration, it's data entry. Someone — usually the maintenance planner or a plant engineer during a slow shift — has to go through every AMC folder and copy vendor name, start and end date, cost and schedule onto the matching asset. For a 40-machine unit with contracts starting and ending on different months, that's a proper one-time exercise, not something to rush through the week before an audit. It's worth doing this alongside a clean pass through your asset registry — coverage checking is only as reliable as the dates sitting behind it.

Where the process still needs a human

The AMC schedule field — monthly, quarterly, half-yearly, yearly — is stored but nothing acts on it. No visit gets auto-generated, no vendor gets nudged. If your contract includes quarterly preventive check-ups independent of breakdowns, you'll still need to plan those separately, perhaps through preventive maintenance built for internally scheduled work. Likewise, there's no AMC invoice or payment schedule tied to the record, so reconciling what you owe a vendor each quarter still happens outside this module, the way it does today.

What this won't tell you

If SLA hours matter to your vendor negotiations — a 4-hour response commitment, say — that field is captured on the call but never checked against actual response time. No breach gets flagged, no penalty computed automatically. There's also no rollup of total AMC spend by vendor, so if your plant head wants to know which of the six vendors servicing those 40 machines is costing the most across the year, that's a number you'll still build outside this screen, not one this module produces for you.

Common ways this goes wrong

• Treating the expiry alert as sufficient on its own: it fires once every 6 days inside a 30-day window, but only reaches people holding breakdown approval rights — if the person who actually renews contracts doesn't have that permission, the alert lands and nobody acts on it.
• Letting a flagged bill pass anyway: the auto-tag on a warranty/AMC-covered bill is a prompt to check chargeability, not a block. Whoever holds service-call approval still has to read it before passing the bill.
• Assuming one AMC entry covers a vendor's whole contract: if a single vendor services 12 machines under one commercial agreement, that's 12 separate asset-level entries, not one.

What's worth watching after rollout

The number worth tracking monthly isn't one this page will invent for you — it's the split between service calls closing as "paid" versus "under AMC" or "under warranty," checked against what your contracts actually promise to cover. That comparison, done by hand against your vendor list, catches drift before it shows up as an unexplained repair bill. It's a smaller habit than a full OEE review, but for plants building out a broader maintenance discipline — the kind covered in what a CMMS actually does or in thinking around total productive maintenance — getting AMC leakage under control tends to be one of the first, cheapest wins.