Out-of-tolerance calibration process map

A swimlane map of what happens when a calibration fails acceptance criteria, from deviation assessment to a closed corrective-action report, across the Calibration Lab, QA, Customer Account, Field Service and ServiceDesk. It starts where the routine calibration map stops, and its whole shape is organised around one hard gate: the customer has to be told before anything is fixed.

Open this map in the editor

Opens a copy in the editor and saves it in this browser. No account, nothing sent anywhere.

What is in this map

15 steps across 5 swimlanes and 5 phases, with 2 decision points.

Swimlanes (who does the work)
Calibration Lab, QA / Documentation, Customer Account, Field Service and ServiceDesk
Phases (left to right)
Deviation & Impact Assessment, Customer Notification, Corrective Action, Recalibration and Documentation & Closeout

Why map this process

An out-of-tolerance instrument is not just a broken part; it is a question mark over every decision that instrument's readings fed since it last passed calibration, and that is what the impact-assessment step exists to answer before anyone reaches for a screwdriver. Cross-referencing the instrument's calibration history against the customer's own batch or production records is what turns 'the reading was off' into a specific, documented list of what may need re-checking.

The notify-before-fix gate is the map's real argument. It would be faster to adjust or repair the instrument first and mention the finding afterwards, and that is exactly the shortcut this map is drawn to prevent: adjustment, repair or replacement does not start until the customer has been told, in writing, what the instrument may have gotten wrong. A customer who finds out after the fact that a fix was already made loses the chance to decide for themselves what needs re-checking on their end.

Every step in the map

This is the spreadsheet behind the diagram. The numbers in “Goes to” are row numbers, which is exactly what the editor’s “Line to” column holds — so you can read the flow here and retype any part of it.

Every step in the Out-of-tolerance calibration (deviation assessment to closeout) process map, in the order the editor numbers them.
#StepShapeSwimlanePhaseGoes to
1 Calibration fails acceptance criteria (*as-found* out of tolerance) Start Calibration Lab Deviation & Impact Assessment 2
2 Assess deviation: magnitude and direction of drift Process Calibration Lab Deviation & Impact Assessment 3
3 Determine impact: which process or product decisions relied on this instrument since its last good calibration Process QA / Documentation Deviation & Impact Assessment 4
4 Impact assessed as **significant?** Decision QA / Documentation Deviation & Impact Assessment 5 (Significant), 6 (Minor)
5 **Escalate** to account manager for urgent customer notification Process Customer Account Customer Notification 7
6 Prepare standard out-of-tolerance notification Process Customer Account Customer Notification 7
7 **Notify customer** of out-of-tolerance finding and impact assessment Email Customer Account Customer Notification 8
8 Determine corrective action: adjust, repair or replace the instrument Process Field Service Corrective Action 9
9 Schedule technician and arrange spares or replacement unit Process ServiceDesk Corrective Action 10
10 Adjust, repair or replace the instrument Process Field Service Corrective Action 11
11 Recalibrate the instrument Process Calibration Lab Recalibration 12
12 Recalibration passes **acceptance criteria?** Decision Calibration Lab Recalibration 13 (Pass), 8 (Fail)
13 Document corrective action and impact assessment Process QA / Documentation Documentation & Closeout 14
14 Share **corrective action report** with customer Process Customer Account Documentation & Closeout 15
15 Out-of-tolerance case closed, instrument returned to service End Customer Account Documentation & Closeout End

The decision points

Every branch in the map, with the label on each outgoing line. These are the questions the process has to be able to answer.

  • 4. Impact assessed as **significant?**

    Owned by QA / Documentation · Deviation & Impact Assessment

    • Significant → step 5
    • Minor → step 6
  • 12. Recalibration passes **acceptance criteria?**

    Owned by Calibration Lab · Recalibration

    • Pass → step 13
    • Fail → step 8

Notes on specific steps

These notes travel with the map. In the editor they live in the Notes column and appear when you open a box.

1. Calibration fails acceptance criteria (*as-found* out of tolerance)
Flagged during the as-found step of the Instrument Calibration & Documentation SOP, before any adjustment is made. That SOP's happy path stops here; this SOP is what runs instead.
3. Determine impact: which process or product decisions relied on this instrument since its last good calibration
Pull the instrument's calibration history and cross-reference the customer's batch, production or release records covering the period since the last good calibration. This is what turns "the reading was off" into a documented list of what may be affected.
7. **Notify customer** of out-of-tolerance finding and impact assessment
Hard gate: adjustment, repair or replacement does not start until the customer has been told, in writing, what the instrument may have gotten wrong and what it was relied on for since its last good calibration.
9. Schedule technician and arrange spares or replacement unit
Confirm whether the fix needs a site visit, a part from stock, or can be handled by exchanging the instrument for a freshly calibrated spare.
13. Document corrective action and impact assessment
Corrective action and impact assessment are recorded together on the case file — an auditor reading it later needs to see not just what was fixed, but what was affected while the instrument was wrong.

Making it yours

The significance decision after impact assessment is one Decision row with two branches — 'Significant' escalates to the account manager for urgent notification, 'Minor' goes to a standard notification — and both branches rejoin at the same customer-notification row before anything else happens. When adapting this map for your own severity thresholds, keep that rejoin: the two paths differ in urgency and who is on the phone, not in whether notification happens at all.

What to watch out for

  • Do not let the recalibration-fails loop skip back past corrective action. The map routes a failed recalibration (row 12, 'Fail') straight back to 'determine corrective action', not to notification or scheduling — the customer has already been told once; a second round of the same fix does not need a second notification unless the assessment itself changes.
  • Documentation and impact assessment belong on one record, not two. Row 13 explicitly pairs 'document corrective action' with the impact assessment from row 3, because an auditor reading the case file later needs to see what was fixed and what was affected while the instrument was wrong, in one place.
  • Do not treat 'schedule technician and arrange spares' as optional for a straightforward adjustment. This map gives it its own row even when the fix looks minor, because confirming whether the job needs a site visit, a stocked part, or a swap for a calibrated spare is what prevents a technician showing up to a job they cannot finish.

Frequently asked questions

Does every out-of-tolerance finding go through the 'significant impact' escalation path?

No — the map branches on the impact decision at row 4 specifically so a minor deviation with no real consequence downstream does not consume the account manager's time the way a significant one does. What is shared by both branches is that the customer is notified either way; only who notifies them and how urgently changes.

What if the customer wants the instrument repaired before they've been formally notified?

The map treats notification as a hard gate specifically because that request happens. The corrective-action decision sits downstream of the notification row, not upstream of it, so there is no branch that lets a fix start first — informing the customer in writing of what may have gone wrong is what the rest of the corrective action is accountable against.

How is this different from a normal repair ticket?

A repair ticket asks 'what's broken and how do we fix it'. This map exists because a failed calibration raises a second question a repair ticket never asks: what did this instrument's bad readings already affect, and who needs to know before the fix happens. That impact assessment and the notification gate built around it are what make this its own process.

Open the map and start editing

The editor loads this chart with the spreadsheet underneath it. Change a cell and the diagram redraws — no drawing, no signup.