Home / Glossary / Recurrence detection

What is recurrence detection in manufacturing?

~3 min read Quality management

Illustration of a manufacturing and continuous improvement glossary

Definition

Recurrence detection is the practice of monitoring a corrected defect or nonconformity to confirm it does not recur. It tracks the same failure mode over a defined window after a corrective action closes, and flags any repeat so the action can be reopened and the root cause revisited.

Recurrence detection is the feedback half of any corrective action. Without it, a defect is treated as fixed the moment the action is signed off, even though the only proof a fix worked is that the problem stops returning.

1How does recurrence detection work?

Recurrence detection turns a closed corrective action into a monitored hypothesis. The team defines what a repeat looks like, watches for it over a set period, and acts the moment the failure mode returns. It runs in four steps:

  1. Define the failure mode. State precisely what counts as a recurrence: the same defect code, part number, process step and root cause, not just any defect on the line.
  2. Set a monitoring window. Choose a time period or a production volume over which the failure mode must not reappear before the fix is considered proven.
  3. Track the signal. Watch the defect rate or the metric tied to the original problem across the window, ideally from live production data rather than a one-off recheck.
  4. Trigger on a repeat. If the failure mode returns inside the window, flag it, reopen the corrective action, and treat the recurrence as evidence the true root cause was missed.

2Why does recurrence detection matter for manufacturing?

Repeat failures are expensive and erode confidence in the quality system. A defect that returns after a documented fix means scrap, rework and, often, an escaped part reaching the customer. Recurrence detection is what separates a problem that is closed on paper from one that is closed in practice.

ISO 9001:2015 clause 10.2 requires organisations to review the effectiveness of any corrective action taken. Detecting recurrence is the practical test of that effectiveness. It is the monitoring step inside a closed-loop improvement system, and the gate that decides whether a CAPA can truly be closed or must be reopened.

3Recurrence detection vs effectiveness verification

The two terms are often used interchangeably, but they answer different questions. Effectiveness verification is a check made shortly after the fix: was the action implemented as planned and did it produce the intended result? Recurrence detection is the longer watch that follows: over time, does the problem stay gone? A fix can pass verification and still fail recurrence detection weeks later, which is the signal that the root cause was never addressed.

Effectiveness verification Short term

A point-in-time confirmation made soon after the action closes. It asks whether the fix was carried out correctly and had the expected immediate effect on the defect.

Recurrence detection Ongoing

A sustained watch across a defined window. It asks whether the same failure mode comes back, and reopens the action if it does, proving the root cause was genuinely removed.

References

  1. International Organization for Standardization, ISO 9001:2015 clause 10.2 Nonconformity and corrective action. iso.org/standard/62085.html
  2. American Society for Quality, Corrective Action. asq.org/quality-resources/corrective-action
  3. Lean Enterprise Institute, PDCA. lean.org/lexicon-terms/pdca
  4. Automotive Industry Action Group, CQI-20 Effective Problem Solving Guide. The reference describing verification of corrective action and monitoring for recurrence in the 8D and Global 8D problem-solving process.