How to Manage Composite UT and Thermography Data in an Inspection ERP

Managing composite inspection data in an ERP means capturing ultrasonic C-scan images, flash thermography sequences, acceptance criteria, and disposition decisions for every laminate panel or bonded structure in a single traceable record, rather than storing raw scan files on a technician's laptop and a separate paper disposition sheet in a filing cabinet — a gap that becomes unmanageable the moment a composite shop scales beyond a handful of parts per week.

Why Composite Inspection Data Is Harder to Manage Than Metallic

Composite nondestructive testing under ASTM E2533 (the standard guide for quantitative NDE of polymer matrix composites) produces fundamentally different data than a metallic UT thickness reading. A single phased array C-scan of a wind blade shear web or an aircraft wing skin can generate a multi-megabyte image file mapping porosity, delamination, and disbond indications across the entire scan area, not a single numeric reading. Flash thermography per ASTM E2582 adds a time-resolved thermal image sequence on top of that. Acoustic emission monitoring under ASTM E2661 during proof loading adds a third data type entirely: time-series event data correlated to load. A spreadsheet built for metallic UT thickness logs simply has no field structure for any of this, which is why composite shops that try to manage inspection records this way end up with scan images scattered across shared drives, disconnected from the disposition decision that was actually made against them.

The Acceptance Criteria Problem

Unlike API 510 or ASME Section V acceptance criteria for metallic components, composite acceptance criteria are almost always customer- or program-specific, referencing an engineering drawing note or a structural repair manual allowable rather than a universal code table. A single shop might be running composite panels against a dozen different customer specifications simultaneously, each with different indication-size thresholds for porosity percentage, delamination area, and bond-line void content. Getting the wrong acceptance criteria applied to a scan — using an outdated revision or the wrong customer's allowable — is one of the most common root causes of composite quality escapes, and it is almost entirely a document control failure rather than an inspection technique failure.

What Belongs in a Composite Inspection Record

• The raw C-scan or thermography image file, geolocated to the exact panel zone or fastener grid location.
• The specific customer specification and revision the scan was evaluated against.
• Indication mapping: type (porosity, delamination, disbond, foreign object inclusion), size, and location relative to structural reference geometry.
• The certified Level II or III technician's disposition: accept, reject, or refer to engineering for a repair evaluation.
• Any subsequent engineering disposition and repair scheme reference if the part was not accepted as-is.
• Calibration records for the phased array or thermography instrument used, tied to the specific scan date.

Centralizing Records in an ERP Instead of Shared Drives

Composite manufacturers and MRO shops that consolidate this data inside a purpose-built inspection ERP get two things a shared drive can never provide: a searchable link between the raw scan file and the disposition decision, and an audit trail showing exactly which acceptance criteria revision was in effect when the disposition was made. This matters enormously during a customer quality audit or a warranty investigation years after delivery, when the shop needs to reconstruct exactly what was inspected, against what standard, and who made the call — a reconstruction that takes minutes from a structured record and can take days from a folder of loose scan files.

Document Control Is the Real Bottleneck

Most composite quality escapes traced back to inspection are not scan interpretation errors; they are document control failures — an inspector evaluating a valid scan against a superseded revision of the acceptance criteria, or a customer-specific specification update that never propagated to the shop floor. A dedicated document control module that locks the current specification revision to the part number and blocks report generation against an outdated revision closes this gap structurally, rather than relying on an inspector to remember to check for updates manually before every scan.

Integrating Multiple NDT Methods on One Part

Composite structures routinely require more than one inspection method on the same part — a UT C-scan for internal porosity and delamination, followed by tap testing or thermography for near-surface disbond, and occasionally acoustic emission during proof load testing. Managing these as separate paper trails makes it easy to lose track of whether every required method was actually completed before final release. An NDT-specific ERP platform that tracks required inspection methods per part number as a checklist, rather than relying on traveler sign-off alone, prevents a part from shipping with one required method still outstanding — a failure mode that is far more common on composite parts than on simpler metallic components precisely because composite acceptance often depends on more than one method to fully characterize a defect.

Traceability Across the Structure's Service Life

For long-life composite structures — aircraft primary structure, wind turbine blades, pressure vessels — the original manufacturing inspection record needs to remain retrievable decades later, correlated against in-service inspection findings at the same location, to support fitness-for-service decisions as the structure ages. Composite shops that build this traceability into their ERP from the start avoid the far more expensive problem of trying to reconstruct manufacturing-era inspection data from archived paper records when an in-service finding raises a question about original build quality.

Practical Recommendations

Composite fabricators and MRO facilities scaling beyond ad hoc scan storage should prioritize three things: geolocating every scan to a structural reference system so findings are spatially comparable over time, locking acceptance criteria revisions to part numbers so the correct standard is always applied automatically, and linking every inspection record to its disposition and any subsequent repair so the full lifecycle history stays intact. Getting these three fundamentals into a structured system early prevents the far costlier retrofit of trying to organize years of accumulated scan files after a customer audit or a fleet-wide quality investigation forces the issue.