Sign in

Blog · Territory, capacity and quota planning · Supply chain

OTIF on ten purchase order lines: the whole arithmetic on one page

The complete on-time-in-full calculation on ten purchase order lines from two suppliers to two sites, small enough to check by hand: the confirmed due date and the tolerance window, the receipt date and quantity, on time and in full per line, the over-receipt that fails the assertion, OTIF per supplier and per supplier-site, why the supplier figure hides the site, the three-month trend that makes a watch list, and the identity that receipts join to lines, so a reader can reproduce every figure and then run it on their own exports.

The short answerTen purchase order lines, two suppliers, two plants. On time is receipt within a window of two days early to zero late against the confirmed due date; in full is received quantity equal to ordered within a stated tolerance. Per line, both must hold. One line is over-received and fails the assertion that receipts never exceed the order; it is listed and excluded. Supplier S-1 is at 83 percent OTIF overall, 100 percent at Plant A and 67 at Plant B. The supplier scorecard shows 83; the site view shows the problem. Every number can be reproduced by hand.

On-time-in-full is a join of two exports and two tests per line, and on ten lines it can be done by hand. This page works it: the window, the receipt, the two tests, the over-receipt, OTIF per supplier and per site, the trend, and the identity.

The rules

On time: received between 2 days early and 0 days late against the confirmed due date In full: received quantity within ±2% of ordered (exact under 50 units) OTIF line: both hold Assertion: Σ received per line ≤ ordered

The ten lines

Line Supplier Site Ordered Confirmed due Received Qty received On time In full OTIF
1 S-1 A 500 3 Sep 2 Sep 500 Yes Yes Yes
2 S-1 A 200 5 Sep 5 Sep 198 Yes Yes (99%) Yes
3 S-1 A 1,000 8 Sep 8 Sep 1,000 Yes Yes Yes
4 S-1 B 500 4 Sep 9 Sep 500 No (5 late) Yes No
5 S-1 B 300 6 Sep 6 Sep 270 Yes No (90%) No
6 S-1 B 800 10 Sep 9 Sep 800 Yes Yes Yes
7 S-2 A 400 2 Sep 1 Sep 400 Yes Yes Yes
8 S-2 A 150 7 Sep 12 Sep 150 No (5 late) Yes No
9 S-2 B 600 9 Sep 8 Sep 660 Yes Over: assertion fails Excluded
10 S-2 B 250 11 Sep 11 Sep 250 Yes Yes Yes

Line 9: 660 received against 600 ordered. The assertion fails; the line is listed as an over-receipt, probably a receiving error, and excluded from OTIF until corrected.

OTIF per supplier and per site

Supplier Site Lines counted OTIF lines OTIF
S-1 A 3 3 100%
S-1 B 3 1 33%
S-1 All 6 4 67%
S-2 A 2 1 50%
S-2 B 1 (line 9 excluded) 1 100%
S-2 All 3 2 67%

Both suppliers at 67 percent overall. S-1 is perfect at Plant A and failing at Plant B. The supplier scorecard says 67 and nothing else.

The trend, S-1 at Plant B

Month Lines OTIF
June 22 89%
July 24 80%
August 21 71%
September 3 so far 33%

Three consecutive declines from a stable base. Watch list, with the trailing three months' spend at that site beside it.

The identity

every receipt joins to exactly one PO line: 10 receipts, 10 lines, joined Σ received ≤ ordered per line: 9 pass; line 9 listed

Where it goes wrong, even at ten

Requested date used. If line 3 had been requested for 1 Sep and confirmed for 8 Sep, it reads seven days late against a date the supplier never agreed.

Supplier level only. S-1 at 67; Plant B at 33 invisible.

Over-receipt counted as in full. Line 9 inflates S-2 at Plant B.

One month escalated. August alone is a bad month; June to September is a deterioration.

From ten to ten thousand

The same two tests per line, per supplier-site-month, with the trend rule. Covirage runs it on the PO and receipt exports every month. The OTIF trend guide covers the watch list, and the lead time variability guide covers the companion measure from the same receipts.

Questions people ask

Why the confirmed due date?

Because the requested date is the buyer's wish and the confirmed date is the supplier's promise. Measuring against the request marks a supplier late for a date it never agreed to, and the supplier stops reading the scorecard.

What is the in-full tolerance?

Stated: here plus or minus two percent of ordered quantity, or exact for lines under fifty units. A line at 98 percent of the order is in full; one at 90 is short. The tolerance is on the report and the same for every supplier.

How does one bad month become a watch list entry?

It does not. Three consecutive monthly declines from a stable base, or a single fall past a stated size, put a supplier-site on the list, ranked by spend at risk. The page shows the three months for S-1 at Plant B.