A formal CAPA in aerospace should be triggered based on risk, repeatability, and potential impact, not just the existence of a single non-conformance report (NCR). Most organizations use defined, documented criteria aligned with AS9100 and customer contracts, and then apply those criteria consistently across products, programs, and sites.
Typical risk-based triggers for formal CAPA
While exact criteria should be defined in your QMS, a non-conformance commonly escalates to formal CAPA when one or more of the following are true:
- Safety or airworthiness risk: The non-conformance could affect flight safety, reliability, or airworthiness, even if it was detected before release.
- Product in the field: The issue is found on delivered hardware, spares, or in-service assets, or could exist in already delivered product due to shared processes or tooling.
- Regulatory or customer requirement impact: The event suggests potential noncompliance with AS9100/AS9110/AS9120, airworthiness directives, customer-specific specs, or contractual quality clauses.
- Repeat or trend: The same or similar non-conformance recurs, or trend data (NCRs, MRB dispositions, escapes, concessions) show a pattern that indicates a systemic issue.
- Systemic process weakness: Evidence that a process, tooling, software system, or control plan is fundamentally ineffective (for example, inspection plans consistently missing a defect mode).
- Critical characteristic or key characteristic impact: The non-conformance involves safety-critical features, key characteristics, or special processes where a single failure mode can have disproportionate impact.
- Significant customer, regulatory, or internal stakeholder concern: A customer complaint, audit finding, or regulatory observation identifies the issue as significant or systemic, not a one-off mistake.
- High cost of poor quality or disruption: The event drives substantial scrap, rework, re-inspection, schedule slip, or AOG risk, suggesting structural process or planning issues.
In practice, many aerospace QMSs use a risk scoring or classification scheme (for example, critical/major/minor) and define CAPA thresholds by class, recurrence, and trend.
Cases that usually do not require formal CAPA
Not every NCR should lead to a CAPA. Escalating everything overloads engineering and quality and dilutes focus on genuine systemic risks. Typical cases where a formal CAPA is usually not required, if they remain isolated and low risk, include:
- Truly one-off human errors with clear local correction and no indication of broader process weakness.
- Minor non-conformances that do not impact fit, form, function, safety, airworthiness, or regulatory/customer requirements, and are cleanly contained and corrected.
- Known, controlled issues already covered under an active CAPA, waiver, deviation, or concession with defined controls and monitoring.
Even in these cases, the NCR should still be logged, evaluated, and trended, because patterns can emerge over time.
Key criteria to formalize in your QMS
To make CAPA triggers defensible and auditable, most aerospace organizations explicitly define:
- Risk categories: How you classify non-conformances (critical/major/minor, A/B/C, or similar) with clear examples.
- Escalation rules: For example, any critical non-conformance triggers CAPA; a defined number of similar majors in a period triggers CAPA; or specific customers or regulators require CAPA for certain types of events.
- Trend thresholds: How repeated minor issues or rising counts of related NCRs drive escalation, even if individual events are low risk.
- Field vs in-process: Stricter triggers for in-service or delivered product versus in-process catches, especially where escape potential exists.
- Special processes and key characteristics: Lower tolerance for issues touching special processes, safety critical features, or serialized, life-limited parts.
These rules should be documented in your QMS procedures, trained, and consistently applied across sites, not improvised case by case.
Brownfield reality and system coexistence
In a typical aerospace environment, CAPA triggers are split across multiple systems: legacy QMS, MES, ERP, PLM, and sometimes separate NCR or MRB tools. This creates gaps if:
- NCRs in MES are not consistently linked to CAPA records in the QMS.
- Suppliers use separate NCR and 8D/RCCA tools, with limited digital feedback into your internal systems.
- Trend detection relies on manual exports or spreadsheets, so systemic issues are noticed late.
When defining CAPA triggers, it is important to be explicit about where risk scoring occurs, how trends are detected across systems, and how traceability between NCR, MRB, deviation, concession, and CAPA records is maintained. In validated or regulated environments, full replacement of QMS or MES tooling is often constrained by qualification burdens and downtime risk, so you may need to layer analytics and workflow on top of existing systems rather than replacing them outright.
Practical governance practices
To keep CAPA triggers effective without overloading the organization:
- Use a cross-functional review board (for example, MRB or quality review board) to confirm which NCRs need CAPA, based on defined criteria.
- Differentiate: Reserve formal CAPA for truly systemic or high-risk issues and use lighter corrective action or local containment for isolated, low-risk events.
- Track linkage: Ensure each CAPA lists the NCRs, concessions, and potentially affected work orders or serial numbers, so that traceability is audit-ready.
- Revisit thresholds: Periodically review if your CAPA trigger rules are too aggressive (backlog, shallow investigations) or too loose (late discovery of systemic issues).
Ultimately, an aerospace non-conformance should trigger a formal CAPA when the evidence shows (or strongly suggests) a systemic, repeatable, or high-risk issue that cannot be adequately managed by local correction and containment alone.
