The planning physics
Available-to-promiseATP
The dated quantity of eligible, uncommitted supply the business can still assign to a new order under its current promising rules — not the same as on-hand inventory.
Operator definition
A customer does not need to know how much inventory the company owns. The customer needs to know that promised supply in the requested quantity, at the requested place, on a date the operation can actually support, will actually come through.
ATP converts supply state into that commercial answer. It considers the supply the promising policy admits — often eligible on-hand inventory and dated receipts — then subtracts accepted demand, reservations, allocations, or protected quantities that have already consumed the same future. The exact buckets vary by system; the durable concept does not: a new promise must not reuse supply that another promise already owns.
Why it matters
A false promise can look like a sales win until the same units are expected by two customers. ATP gives order takers a forward-looking commitment boundary. It lets the business say yes faster because the answer is grounded in dated supply and existing obligations rather than a warehouse total or a planner's memory.
A strong ATP process also creates discipline between planning and execution. It forces the organization to decide which supply types count, how forecasts consume future availability, which customers or programs receive protection, and when a confirmation becomes binding.
The planning physics
Systems differ on forecast consumption, receipt categories, backward or forward consumption, priority, product allocation, time buckets, substitutions, and whether late or low-confidence supply is admitted.
This is not a universal formula. The physics is the conservation of a claim: once a unit of dated supply is committed, the same unit cannot safely support a second commitment unless the first claim is cancelled, superseded, or deliberately reallocated.
Worked example — the promise waterfall
Change any input to see eligible supply and ATP recompute. Defaults to a site promising through July 10.
| Step | Δ | Balance | |
|---|---|---|---|
| On hand | +1,000 | 1,000 | |
| On quality hold | −100 | 900 | |
| Scheduled receipt | +500 | 1,400 | |
| Accepted commitments | −1,100 | 300 | |
| Protected quantity | −100 | 200 | |
| Available-to-promise | 200 |
The remaining short quantity is not a "no" — it is the question CTP answers: can new supply be made, bought, moved, or expedited in time?
What goes wrong without it
- Sales promises against gross on-hand inventory while operations knows some stock is blocked or reserved.
- Concurrent users consume the same inbound receipt because availability checking and reservation are separate transactions.
- An order is confirmed against a receipt whose date or confidence has already deteriorated.
- Aggregate ATP is positive, but the supply is in the wrong location, wrong lot state, wrong configuration, or wrong priority class.
- A new planning run silently reassigns supply that already supported a customer confirmation.
The result is not merely a late shipment. It is a broken commitment model.
How it shows up in high-consequence supply chains
In high-consequence settings, promiseability may depend on more than item and date. A blood component must be compatible and sufficiently fresh. A vaccine lot must remain viable and released. A clinical-trial kit may be country- and protocol-specific. A critical spare may be qualified for one asset configuration but not another.
Available is a demand-relative judgment, not a global inventory flag. The same physical units can be promiseable to one demand and not another by identity, location, quality, expiry, or priority class:
| Demand class | Eligible units | ATP | Why |
|---|---|---|---|
| Emergency protected | 1,400 | 300 | May draw on the 100 protected units |
| Standard order | 1,400 | 200 | Protection withheld |
| Trial kit, country locked | 300 | 0 | Only protocol- and country-eligible lots count |
In a blood, vaccine, or clinical-trial context, ATP must be calculated against the relevant identity · location · quality · expiry · priority · commitment states.
Common confusion: ATP vs CTP
ATP is not on-hand inventory. On-hand may be blocked, reserved, expired, inaccessible, or protected.
ATP is not CTP. ATP asks what existing eligible supply remains promiseable. CTP asks what additional supply can feasibly be made, bought, moved, or expedited.
| Available-to-promise | Capable-to-promise | |
|---|---|---|
| Asks | What existing eligible supply remains promiseable? | What additional supply can feasibly be created in time? |
| Looks at | On-hand plus dated receipts, minus commitments and protection | Make, buy, move, or expedite within constraints and lead time |
| Answers | A dated quantity you can commit now | A feasibility and date for supply that does not yet exist |
| In the example | Covers 200 of the 300-unit request | Tests whether the remaining 100 can be created by July 10 |
ATP is not certainty. A positive result is only as reliable as the state, dates, policies, and commitments behind it.
Vista point of view
ATP should be explainable as a supply-and-commitment trace, not merely displayed as a number. A green availability result is not enough if the operator cannot see which supply elements support the promise, which claims were subtracted, which policy protected supply, which run calculated the result, and what authority created the final commitment.
Vista's view is that ATP is a promise discipline. It is where planning state becomes a customer-facing answer: yes on the requested date, yes for a partial quantity, yes through a different source, or no until CTP finds a feasible path. The better the trace, the faster the organization can act without turning every order into a manual reconciliation.
The upside is commercial as well as operational. A strong ATP process helps teams protect scarce supply for the right demand, enforce customer or program priorities, detect deteriorating receipts before they become broken promises, and know when to escalate from ATP to CTP.
That is confidence by construction. The order is accepted because the future has been accounted for, not because a screen showed green. If ATP cannot support the full request, the operator should know exactly which assumption, constraint, or commitment closed the option.
Sources Reviewed 22 June 2026
- SAP Learning describes ATP checks as cumulative availability calculations on the material availability date; confirmed requirements reduce ATP quantity, and the scope of check determines which stocks, receipts, and requirements are included: Available-to-Promise basics ↗.
- Oracle Global Order Promising documents ATP rules as instructions that tell Promising what supply to search — on-hand supply, purchase orders, and work orders — and how rules can vary by item or organization: Available-to-Promise Rules ↗.
- ASCM/APICS treats discrete and cumulative available-to-promise, with capable-to-promise, as part of customer order promising and order entry in its CPIM body of knowledge: CPIM Exam Content Manual ↗.
- Microsoft Dynamics 365 is supporting implementation evidence for ATP as order-promising logic based on uncommitted inventory, receipts, issues, and lead-time controls: Order promising and ATP ↗.
- Exact calculation and protection rules are implementation-dependent; the canonical definition above is a cross-implementation synthesis.