A pulverised-sample screen check fails its required passing criterion. The laboratory identifies the nonconforming work, stops the affected preparation stream and isolates the material. Containment may be completed within hours. Corrective action and effectiveness review may take longer.
But neither event answers the final question: did the action prevent the problem from recurring?
That distinction matters in mining and metallurgical laboratories. A revised procedure, serviced equipment or completed retraining confirms that an action was implemented. It does not establish that the action prevented recurrence.
OnQMS provides a controlled path from the original laboratory finding through a nonconformity report (NCR) and the corrective branch of a corrective and preventive action (CAPA) record to documented effectiveness verification. Laboratory personnel identify the nonconformity, determine the root cause and judge whether the evidence supports closure. OnQMS keeps the records, responsibilities, dates and decisions connected so that path can be reviewed.
The failed screen check comes before the NCR
Consider a routine check on pulverised material. The retained fraction exceeds the laboratory's or contract's applicable criterion. The laboratory—not the software—recognises that the preparation result is unacceptable and must determine the extent of the problem.
The first response is containment. The affected work may need to be stopped and segregated while the laboratory establishes which samples, batches or reported results could be involved. Containment limits immediate exposure; it does not establish why the failure occurred or prevent it from happening again.
A useful nonconformity report needs to preserve the decisions made at this stage. In OnQMS, an NCR can record:
- the significance of the event;
- the containment applied;
- the disposition of affected work;
- whether the client was notified;
- whether resumption of work was authorised; and
- the site responsible for the record.
These fields do not make the decisions. They preserve what authorised laboratory personnel decided under the control of nonconforming work in ISO/IEC 17025 §7.10 and create a traceable basis for corrective action under §8.7.
Before corrective action, determine what may have been affected
A failed screen check does not necessarily affect every sample prepared by the laboratory. It may relate to a specific pulveriser, interval, batch, material type, operating condition or maintenance state. The laboratory must define the scope using the evidence available.
That assessment may require questions such as:
- When was the last acceptable screen check?
- Which samples were prepared after that check?
- Were any results authorised or issued during the potentially affected interval?
- Does the material require re-preparation or reanalysis?
- Is client notification necessary?
- What evidence is required before preparation resumes?
This is where a simple “issue closed” status becomes dangerous. If significance, containment, disposition and authorisation are not preserved, the organisation may later know that an event occurred without being able to demonstrate how the possible impact was controlled.
OnQMS keeps the NCR active through a defined lifecycle: Open, Investigating, In Action, Verifying and Closed. Active records remain visible; overdue records are distinguished from closed ones. Open NCRs with deadlines can participate in alerts, overdue tracking and email summaries, reducing the risk that a quality event disappears into an inbox or spreadsheet.
Containment is not corrective action
The distinction is operational:
| Step | Question answered |
|---|---|
| Containment | What must we control immediately? |
| Correction or disposition | What will we do with the affected work? |
| Root-cause analysis | Why did the event occur? |
| Corrective action | What will remove or control the cause? |
| Effectiveness verification | What evidence shows that the action worked? |
In the screen-check example, cleaning the pulveriser and reprocessing the affected material may correct the immediate condition. That response may be necessary, but it is not automatically a complete corrective action.
The investigation may identify a worn component, unsuitable cleaning frequency, an incorrect setup, ineffective inspection, an unrecorded maintenance intervention or another cause supported by the laboratory's evidence. Laboratory personnel record the root cause; OnQMS does not determine it.
Once the laboratory has established a defensible cause, the CAPA can define the plan, assign a responsible person and set a due date. The CAPA remains connected to its originating NCR, preserving the path back to the laboratory event that required action.
Completing the action does not resolve the problem
Suppose the investigation attributes the failed screen check to wear that was not being identified early enough. The laboratory updates the inspection procedure, services the equipment and changes the preventive-maintenance trigger.
Completing those actions establishes implementation, not effectiveness. ISO/IEC 17025 §8.7 requires the laboratory to review the effectiveness of corrective action. Depending on the action and the laboratory's own criteria, the review might use subsequent screen-check results across an appropriate operating period, repeat-event history, maintenance observations or another technically justified measure. The laboratory decides what evidence is suitable and how much is needed.
The central test is not:
Was the action performed?
It is:
Did the action achieve its intended result without the nonconformity recurring under the conditions evaluated?
The answer needs an attributable record, not only a changed status.
What documented effectiveness looks like
OnQMS CAPA records include the effectiveness result, who verified it and when the verification occurred. This creates three distinct pieces of evidence:
- The evaluation: what the laboratory concluded about the action's effectiveness.
- The verifier: the person accountable for that conclusion.
- The verification date: when the evaluation was made.
The CAPA lifecycle—Open, In Progress, Verifying and Closed—keeps verification visible as a separate stage rather than treating closure as the automatic consequence of completing the action plan.
If the evidence does not support effectiveness, the CAPA should not move to closure. The laboratory must continue its investigation, revise the action or define further work. An open CAPA with an unresolved effectiveness review is more credible than a closed record that cannot explain why closure was justified.
Closure is a controlled decision
Quality records need both accountability and segregation of duties. OnQMS separates permission to close NCR and CAPA records from general permission to edit them. This allows the laboratory to configure who may record or update the work and who is authorised to approve closure.
That separation does not prescribe the laboratory's organisation chart, but it supports a controlled workflow. The person implementing an action does not have to be the person who verifies its effectiveness or authorises closure.
NCR and CAPA reports provide retrievable records of the process, and their PDF outputs include electronic-signature manifests. A hash-chained audit history retains previous and updated values. The history is tamper-evident, not tamper-proof.
What an assessor should be able to follow
An assessor reviewing the closed CAPA should be able to move backward through the evidence:
Authorised closure
← effectiveness result, verifier and date
← corrective-action plan, owner and deadline
← recorded root cause
← NCR significance, containment and disposition
← original failed screen check
That path answers more than “was there a CAPA?” It shows why the record existed, how the laboratory controlled the immediate problem, what it concluded about the cause, who owned the action and what evidence supported closure.
The same logic applies beyond sample preparation. An out-of-tolerance intermediate equipment check, an unsatisfactory proficiency-testing result, a complaint or an appeal may also lead to an NCR and corrective action. The technical evidence will differ, but the control principle remains the same: the original finding, the laboratory's decisions and the effectiveness conclusion should remain connected.
Where OnQMS stops
OnQMS organises evidence. It does not certify compliance with ISO/IEC 17025, guarantee accreditation or replace the judgement of laboratory personnel and assessors. A completed CAPA field does not establish effectiveness unless the laboratory has performed and documented an appropriate review. A signature establishes attribution, not technical correctness.
Your laboratory already generates the evidence. OnQMS organises it and shows you what is missing.
This extends the broader evidence architecture described in A Result Is Not Evidence: How OnQMS Builds the ISO/IEC 17025 Evidence Chain. Here, the focus is the specific path from nonconforming work to corrective action and evidence-supported closure.
Test one closed CAPA
Select one CAPA currently marked closed and trace it backwards.
Can you identify the original finding? Can you see the NCR's significance, containment and disposition? Is the root cause attributable to a responsible person? Does the action have an owner and deadline? Most importantly, can you show the effectiveness result, who verified it and when?
If the record ends at “action completed,” the corrective-action process may be administratively finished but technically incomplete.
Related reading
- ISO 17025 LIMS for Mining Laboratories
- OnQMS for ISO/IEC 17020 Inspection Bodies
- A Calibration Certificate Is Not Enough: Intermediate Checks Between Calibrations
Primary reference
International Organization for Standardization. ISO/IEC 17025:2017 — General requirements for the competence of testing and calibration laboratories.