Table of Contents
A supplier can show 100% on-time delivery while one of three order lines remains short. The discrepancy may come from the metric definition rather than an arithmetic error: one report counts the first arrival, another counts complete lines, and a third counts individual units.
For electronic components, a useful scorecard makes that definition visible. On-time delivery (OTD) and on-time in-full (OTIF) should connect to the purchase-order commitment and production need, with consistent rules for split shipments, changing promises and inspection holds.
How do you calculate supplier OTD without changing the denominator?
Define the scoring unit and due-date cohort before calculating the percentage. For a line-based completion measure, divide the order lines completed within the agreed delivery window by all eligible lines due in that reporting period, including unresolved overdue lines.
This avoids a common reporting problem: counting only receipts that happened during the month makes late, still-open orders disappear from the denominator. A receipt-based report can still be useful, but it answers a different question and should be labeled accordingly.
The Oracle JD Edwards supplier-delivery documentation describes configurable early and late delivery tolerances. The important procurement implication is that the window is a policy input. There is no reason to assume that two supplier reports using different tolerances are comparable.
Record the relevant event as well: supplier dispatch, arrival at the receiving dock and release for production are different milestones. A ship-date commitment should not silently become a receipt-date measure.
Why can the same deliveries score 100%, 96% or 66.7%?
Different counting rules weight the same delivery events differently. A first-arrival line measure can treat a partial shipment as on time, a quantity measure weights each unit, and a full-line measure requires the complete agreed quantity.
Consider this hypothetical cohort of three lines, all due on the same date. Assume no early window, no quantity tolerance and no later-approved requirement changes:
| Line | Quantity due | Received by due date | Received late |
|---|---|---|---|
| A | 100 | 100 | 0 |
| B | 100 | 60 | 40 |
| C | 800 | 800 | 0 |
Under these explicitly chosen rules:
- First-arrival line OTD = 3 / 3 = 100%. Every line had some material arrive on time.
- On-time quantity = 960 / 1,000 = 96%. Forty units arrived late.
- Complete-line OTIF = 2 / 3 = 66.7%. Line B was not full by the deadline.
None of these calculations changes the physical receipt history. The metric selects what to emphasize. The 100% figure is misleading if presented as evidence that all requirements were satisfied. Quantity weighting can also conceal the operational importance of a small but critical line.
Which promise date should a supplier scorecard use?
Preserve the original accepted promise and every approved revision, then report the date basis used for each measure. Replacing the original date with the latest promise removes evidence of schedule drift.
A practical ledger keeps the purchase line, requested date, first accepted commitment, revision timestamp, revised commitment, change reason and approving party. It can then show both performance against the original commitment and performance against the latest valid plan.
Buyer-driven changes deserve separate treatment from supplier delays. If the customer postpones an order, retain that event so the report can distinguish a changed requirement from a missed commitment. Agree the treatment before the period closes; adjusting rules after seeing the score undermines comparability.
For planning future receipts, use the component lead-time workflow. A delivery score measures past execution under its defined rules; it does not turn a quoted lead time into a guaranteed production date.
Should a quarantined receipt count as on time?
It may count as an on-time physical receipt under a dock-arrival metric, but it should not count as production-ready material until the required release conditions are met. Keep those events separately visible.
Record the quantity received, quantity on hold, reason for the hold and release date. That permits a delivery measure and a material-availability measure to coexist without hiding either logistics performance or quality disruption.
Finish the scorecard with exceptions that require action: overdue critical lines, recurring short shipments, repeated promise changes and unresolved holds. The percentage provides a consistent trend only when the underlying rule remains stable. The exception list explains which commitments still prevent the next build.
Frequently Asked Questions (FAQ)
Should early deliveries automatically improve a supplier score?
Only under a defined early-delivery tolerance. Material arriving too early can create storage or cash-flow costs, so report early arrivals consistently instead of assuming every early receipt is favorable.
How should an approved order cancellation affect the denominator?
Apply a documented cancellation rule and retain the approval date and reason. Do not remove a late line retrospectively merely because it was canceled after its due date.
Can an average delivery score hide a critical component shortage?
Yes. An aggregate metric can be dominated by high-volume or low-risk lines. Report critical overdue lines and their production consequences alongside the supplier-level percentage.
