Crosschecking Methods and Governance of the NDT Function

Crosschecking Methods and Governance of the NDT Function

In part one of this series, it was set out why independent ISO 9712 Level 3 certified personnel should be engaged when an NDT facility is being audited, particularly in ship and offshore newbuild environments where inspection outcomes can later become contractual, financial, or legal evidence.  The same principle carries directly into the next stage. Once it is established whether a facility appears compliant, we still need a practical way to confirm that the NDT function is delivering reliable results under real production conditions, not just during a scheduled audit.

That is the purpose of crosschecking and ongoing control.  Done properly, crosschecking is not a full re-inspection programme designed to catch people out.  It is a targeted verification activity that answers a simple question:

Is the inspection regime consistently capable of detecting the defect mechanisms that matter, on the welds that matter, in real-world conditions? 

Andrew Waller Principal Inspection Engineer

If the answer is yes, crosschecking reinforces confidence.  If the answer is no, it provides early warning, before the consequences become rework, programme delay, or dispute.

Many quality systems produce paperwork that looks credible. Procedures exist, technicians hold certificates, equipment lists are present, and reports are issued.  But when the pressure rises, as is often the case in production, systems drift.  Calibration habits become informal, scan coverage becomes optimistic in tight access, interpretation trends towards “what usually gets accepted”, and reporting becomes abbreviated to preserve throughput.  This drift is often subtle and rarely malicious. It is simply what happens when technical disciplines are asked to behave like production trades without the same level of technical supervision.

Crosschecking is one of the most effective tools available to detect that drift.  However, the choice of method is critical.  A crosscheck that produces little more than another pass/fail statement adds limited value.  What is really required from crosschecking is objective, reviewable evidence and data that can be interrogated by an independent person later and that can withstand scrutiny if questions arise months or years down the line.  This is why methods such as ACFM and PAUT are particularly well suited to crosschecking in modern fabrication environments.  These methods are capable of creating fully recordable and auditable inspection records.

Why Recordable Crosschecks Change the Conversation

One of the recurring problems in disputes involving NDT is that the argument becomes centred on what someone says they did, rather than what can be proven.  A report may state that a weld was inspected and found acceptable, but if the method used does not produce a retained dataset, it can be difficult to demonstrate later that adequate coverage was achieved, that sensitivity was appropriate, or that interpretation was consistent with the acceptance criteria at the time.

This is where international standards and the concept of independence become more than “nice to have”. ISO 9712 provides a recognised framework for personnel competence and certification however certification is only one part of the picture.  When crosschecking is being used to validate adequacy and particularly where the outcomes may influence acceptance, repair strategy, schedule, or commercial position, there is a strong case for involving independent ISO 9712 Level 3 personnel.  Independence reduces the risk of conscious or unconscious bias, and Level 3 competence ensures that the crosscheck is technically meaningful rather than simply a repeat of the original inspection.

A similar principle sits behind ISO/IEC 17020, which is focused on the operation of bodies performing inspection and places emphasis on impartiality and competence.  While not all NDT departments operate as ISO 17020-accredited inspection bodies, the standard usefully captures the mindset that serious stakeholders expect: inspection must be technically competent, demonstrably controlled, and impartial in its decision-making.  Crosschecking performed by independent, appropriately qualified personnel aligns with that mindset, and it strengthens the defensibility of the NDT function.

ACFM as a Crosscheck for Surface Inspection Adequacy

Surface inspection regimes, often based on VT, MT, or PT, remain fundamental, and in many scenarios, they are entirely appropriate.  The issue is not that these methods are “wrong”.  The issue is that their effectiveness can be strongly influenced by access, surface condition, lighting, cleaning discipline, consumable control, and interpretation consistency.  In a fast-paced yard environment, those variables can change from shift to shift.

ACFM is a highly effective crosschecking tool in this space because it can provide a recordable output that supports later review.  Where MT or PT may result in a written report and perhaps some photographs, ACFM can retain scan data that allows an independent reviewer to assess what was observed and how indications were characterised.  That matters when you are trying to validate whether the surface inspection programme is truly capable of detecting surface-breaking cracking at weld toes, in heat affected zones, or in locations where geometry and surface condition make conventional surface methods more variable.

Using ACFM as a crosscheck is therefore not about replacing MT/PT.  It is about selecting a technique that can verify, with stronger evidence, whether the regime is performing as intended, particularly on critical welds, trend locations, repair welds, or where workmanship issues have been observed.  When crosschecking reveals discrepancies, indications found by ACFM that were not reported by the primary surface method, the correct response is not to argue about individual performance.  The correct response is to examine whether the system controls are sufficient: preparation standards, lighting, consumables, procedure adherence, supervision, and acceptance interpretation.

PAUT As a Crosscheck for Volumetric Inspection Adequacy

The same logic applies to volumetric inspection.  Conventional UT remains a widely used method, and in many cases, it is the correct choice.  However, conventional UT can be vulnerable to technique variability, particularly when access is restricted, when weld geometry is complex, or when procedure essential variables are not properly controlled.  It is also vulnerable to the problem that matters most in disputes: the inspection result may not be supported by a retained dataset capable of independent re-review.

PAUT, when applied properly, provides a significantly stronger evidential trail.  A well-constructed PAUT crosscheck can retain A-scans, S-scans, and associated data that demonstrate coverage, sensitivity, and interpretation in a way that can later be audited.  This matters enormously when the crosscheck objective is to validate not only whether defects are being detected, but whether the original technique truly interrogated the relevant volume and whether acceptance decisions were technically defensible.

Importantly, PAUT crosschecking also allows an independent Level 3 to examine whether the procedure design itself is appropriate: scan plan coverage, focal laws, wedge selection, sensitivity settings, and essential variable control.  In other words, PAUT is not simply a “better UT”.  It is a method that can be used to test the adequacy of the entire volumetric inspection approach, people, procedure, equipment, and discipline while producing data that is auditable.

Crosschecking Must Be Independent, Targeted, and Risk-Based

Crosschecking only works if it is designed to measure normal conditions.  If it becomes predictable, it risks becoming a performance staged for an audience.  If it becomes too broad, it becomes expensive and starts to look like re-inspection.  The most effective crosschecking programmes are selective and risk-based, focusing on welds and details where failure consequences are high, where defect mechanisms are credible, and where production realities introduce vulnerability.

In practice, this often means prioritising primary structure, fatigue-sensitive details, thick-section joints, high-restraint areas, complex nodes and attachments, and repair welds, particularly repeated repairs.  It also means paying attention to trends such as repeat defect types, repeat welders, repeat locations, or repeat shifts.  Crosschecking is at its best when it is used as an intelligent diagnostic tool rather than a blunt instrument.

This is also the point where independence matters most.  If a crosscheck is being used to confirm adequacy, it should ideally be performed or overseen by an independent ISO 9712 Level 3 with the authority and competence to challenge the technical assumptions, not just the workmanship.  Independence is not an insult to the yard, it is a control that protects all parties by ensuring that findings are credible, impartial, and technically grounded.

Control of the NDT function: Making Inspection Defensible, Not Just Complete

Crosschecking is one half of the picture.  The other half is governance, the daily controls that ensure the NDT function stays competent and consistent long after the audit team has left. This is where many facilities struggle.  It is relatively easy to demonstrate a controlled system during an audit. It is harder to maintain that control week after week through schedule pressure, turnover, and changing scopes.

The foundation is record keeping that is built for scrutiny, not convenience.  “A report exists” is not the same as “the inspection is defensible.”  Defensible records allow traceability from weld to drawing, from report to procedure revision, from technician to certification validity at the time of inspection, and from equipment to calibration status.  Where advanced methods are used, particularly ACFM and PAUT, the records should also link the issued report to the retained datasets.  If data files can be separated from conclusions, the system is vulnerable.  The record trail should make it straightforward to show what was inspected, how it was inspected, by whom, with what equipment, to what acceptance criteria, and what evidence supports the conclusion.

As inspection becomes increasingly digital, data integrity controls become just as important as calibration controls.  Facilities should be able to show where files are stored, who can access them, what rules exist around re-issue and superseded reports, and how audit trails are maintained.  This is not bureaucracy.  It is evidential hygiene, and it is exactly the sort of discipline that stakeholders implicitly expect when NDT results are used to make high-value acceptance decisions.

Competence control is the next pillar. ISO 9712 certification provides a structure, but certification alone does not guarantee current competence, especially in specialised applications, difficult access, or rapidly changing project conditions.  Strong departments treat competence as something that must be maintained and verified, not assumed.  That means ongoing training that reflects real project challenges, not generic refreshers, with mentoring and supervision where technicians are developing, and structured updates when procedures, equipment, or acceptance criteria change.

Regular trade tests are a particularly effective but often underused tool.  Practical assessments using representative samples and credible defect types provide direct evidence of capability, and they allow technical leaders to detect drift early.  Trade tests should assess scanning discipline and coverage as well as detection, and they should test interpretation consistency against the applicable acceptance criteria.  Where discrepancies arise, the response should be structured, identify root causes, apply corrective actions (which may include procedure changes, equipment checks, or targeted retraining), and then verify effectiveness with follow-up assessment.

Finally, none of this is stable without technical authority.  A recurring weakness in vulnerable NDT systems is that Level 3 involvement is nominal rather than engaged.  Level 3 input should be visible where it matters: method selection rationale, procedure technical adequacy, essential variable control, acceptance interpretation aligned with code intent, and meaningful involvement in corrective action where crosschecking identifies discrepancies.  This is not simply “signing procedures.” It is technical governance.

The Practical Outcome

When crosschecking and governance are treated seriously, the NDT function becomes more than a production service.  It becomes a controlled evidence-generating discipline.  ACFM and PAUT are not highlighted here because they are fashionable methods; they are highlighted because they can create recordable datasets that strengthen auditability and defensibility.  When those crosschecks are designed and overseen by independent ISO 9712 Level 3 personnel, the credibility of the conclusions increases significantly, and particularly when stakeholders include owners, class, insurers, or legal parties.

Ultimately, the objective is not to assume failure.  The objective is to ensure that if questions arise, the answers are supported by evidence rather than recollection.  That is what good crosschecking achieves, and that is what strong governance protects.

Is your shipyard NDT function fit for purpose?

A shipyard NDT audit from The Lab can help you improve your NDT function and ensure it is consistent, reliable and able to provide NDT under robust and demonstrable controls.

Contact The Lab now to discuss your NDT audit needs

For more information and the latest industry insights, explore The Lab's News and Knowledge Hub...

What Is the Heat Affected Zone? | What Is Dye Penetrant Testing? | What Is Scanning Electron Microscopy?