Skip to content
CorrectiveActionReport.comSupplier CAR Response Desk — home

Root Cause Analysis

Occurrence and Escape Root Cause: Why Every 8D Needs Both

Every defect that reaches a customer has two causes: why it was made and why it got out. This guide covers finding the escape point and recording both in D4.

Rovaryn Digital · · 9 min read

A Contained Burr and Two Unanswered Questions

The burr on the mounting hole was contained the day the customer found it. Their line caught the bracket at assembly, your team pulled the lot and sorted what was left in the building, and the interim response went out on time.

A few days later the question is the one every 8D asks: why did this happen? Your team traces the burr to a worn deburring tool, replaces it, adds a tool-change interval, and writes it up. It looks complete.

Then the customer's SQE writes back with a single line: "How did it get past your final inspection?" Nothing in the 8D answers that. The team explained why the burr was made. It never explained why a burred part was packed, shipped and received.

Those are two separate questions, and they have two separate answers. This article shows how to find both, how to keep them apart, and how to record them in D4 so the response holds up when the customer reads it.

Occurrence and Escape Root Cause: Two Questions, Two Chains

An occurrence root cause explains why the defect was created. It lives in the process that made the part: tooling, setup, material, method, training, or design.

An escape root cause explains why the defect was not detected before it left your control. It lives in the detection system: inspection method, sampling plan, gauge, work instruction, reaction plan, or the point in the flow where checks happen.

Many customer formats ask for both, sometimes under the labels "root cause of occurrence" and "root cause of non-detection" or "escape." Check your customer's form and supplier manual to see how they label it. If your customer's format has only one root cause field, you can still analyze both. Put the second chain in the attachment or the narrative.

Occurrence Escape
Question Why was the defect made? Why was it not caught here?
Where it lives Production process Detection and control system
Typical evidence Process records, tool life, setup sheets Inspection records, control plan, gauge data
Corrective action aims at Stopping the defect from being made Making sure a defect would be caught if it were made

Fixing only the occurrence cause leaves the detection gap in place. The next defect, from a cause you did not anticipate, will pass through the same hole. Fixing only the escape cause adds inspection and leaves the process making bad parts. Neither alone is a real corrective action.

IATF 16949 clause 10.2.3 asks for a documented problem-solving process that includes root cause analysis with its methodology and results, along with containment and systemic corrective actions. In the 2025 AIAG Quality Summit, the IAOB ranked problem solving (10.2.3) as the top IATF 16949 major nonconformance, according to simpleQuE's summary of that presentation. Analyzing only half the problem is an easy way to produce a response that reads as thin.

Finding the Escape Point

The escape point is the earliest step in your process where the defect should have been caught and was not. Finding it is a walk-back exercise.

  1. Start at the customer's detection point. Where did they find it: incoming inspection, line assembly, or in the field? Note it.
  2. List every check between creation and shipment. Include in-process checks, operator self-inspection, final inspection, packing audits, and any lot release step.
  3. Identify where the defect first existed. If the burr is created at the deburring step, every check after that is a chance to catch it.
  4. For each check, ask whether it was designed to detect this defect. Many checks look at a different characteristic. A final inspection that verifies hole diameter does not detect a burr on the hole edge.
  5. Mark the earliest check that should have caught it. That is your escape point. Later checks that also missed it are secondary and worth noting.

A useful companion question is whether the check existed at all. A missing check, a check that exists on paper but not on the floor, and a check that exists but cannot detect this failure mode lead to different corrective actions.

Questions for each candidate check

  • Is the characteristic on the control plan or inspection instruction?
  • Is the method capable of seeing this defect (visual versus gauge, sample versus 100%)?
  • Was the inspector trained and given a clear accept/reject standard?
  • Did the check run on this lot, this shift, this machine?
  • If the check found something, did the reaction plan stop shipment?

Each "no" points to a different branch of the escape chain.

Running a 5 Whys on Each Branch

Run two separate 5 Whys chains rather than one that tries to cover everything. If you want a layout for the chain itself, see a 5 Whys template for Excel. The sketch below shows how the two chains differ for the burr example.

Occurrence chain

  1. Why did the part have a burr? The deburring tool was worn.
  2. Why was the tool worn? It ran past its useful life.
  3. Why did it run past its life? No change interval was defined for this tool.
  4. Why was no interval defined? The tool was added to the cell without a tooling review.
  5. Why no review? The change process does not require one for like-for-like tooling.

Escape chain

  1. Why did a burred part ship? Final inspection did not catch it.
  2. Why not? The inspection instruction checks hole diameter, not edge condition.
  3. Why not edge condition? The characteristic was never added to the control plan.
  4. Why not added? The burr was not identified as a failure mode when the plan was written.
  5. Why not identified? The risk review did not include secondary operations on this part family.

Notice the chains end in different places: tooling change control in one, risk review and control plan content in the other. They also lead to different owners and different corrective actions. Stopping at the first plausible answer is the usual trap, so push each chain until the answer is a system you can change.

A fishbone helps when the occurrence cause is unclear. Sorting candidate causes by category such as machine, method, material, measurement, people and environment gives the team a structure to work through. ASQ lists the fishbone diagram, also called the cause-and-effect or Ishikawa diagram, among its seven basic quality tools. Measurement is a natural home for escape candidates.

Recording Both in D4

D4 is the root cause discipline. When you write it up, give each cause its own labeled entry.

Takeaway: a reader should be able to find one answer to "why was it made" and a separate answer to "why was it not caught," each with evidence.

A workable D4 layout:

  • Occurrence root cause: one clear statement, then the supporting evidence (tool records, process data, a repeat of the failure under controlled conditions).
  • Escape root cause: one clear statement, then evidence (control plan extract, inspection instruction, gauge study, a test that shows whether the check could have detected the defect).
  • Verification of each cause: how the team confirmed it, such as turning the cause on and off, or reproducing the escape with a known-bad sample.
  • Contributing factors: things that made either cause more likely but are not the root.

For a broader view of what goes into this discipline, see the D4 root cause analysis guide. For how a finished response reads end to end, see an automotive 8D report example.

Corrective actions in D5 should map back to each cause. If you list five actions and cannot say which cause each one addresses, the analysis probably has a gap.

What D3 Has to Do With Escapes

The escape analysis also tells you whether your containment was aimed correctly. Containment in D3 exists to keep suspect product from reaching the customer while the root cause work continues. If you know the escape point, you know which checks to strengthen immediately, and which stock to sort and where. The window for containment is set by each customer, so read the current revision of your customer's supplier manual or CSRs. For the discipline itself, see D3 containment actions.

Sub-Tier Suppliers and SCAR Escalation

Sometimes the occurrence cause sits upstream. A purchased part or material arrived out of specification and your process did not reject it. That raises two questions again, one for each side.

  • Your escape: why did your incoming inspection or process controls not catch the sub-tier defect? This is yours to own and fix, regardless of where the defect started.
  • Their occurrence: why did the sub-tier supplier make it? That is the subject of a supplier corrective action request (SCAR) you may send down the chain.

Sending a SCAR to your own supplier does not close your 8D. Your customer is asking what you changed. Keep the two threads separate: your response to the customer carries your escape analysis and actions, and the sub-tier SCAR carries their occurrence analysis. Some customers expect to see evidence that you escalated, so check what yours expects.

In aerospace, the same logic applies in root cause and corrective action work under AS9100. SAE AS13000A, issued by the G-22 AESQ Committee, defines the problem-solving process a supplier uses when responding to a customer request (SAE International). For aerospace-specific context, see RCCA in aerospace.

Tools Commonly Used in D4

If you are wondering what tools are used in 8D D4, the common set is small:

  • 5 Whys for tracing a chain of cause and effect, run once per branch.
  • Fishbone for organizing candidate causes before narrowing.
  • Is/is-not for comparing where and when the problem occurs with where and when it could occur but does not.
  • Process and inspection records as the evidence behind any of the above.

AIAG's CQI-20 guide on effective problem solving is another reference that many automotive quality teams know. Whichever tools you use, apply them to both questions.

Keeping the Record

Both chains, the evidence, and the verification steps are worth keeping together per request. A supplier-side record can hold occurrence and escape analyses next to the containment log and actions, so the history of one customer's request is easy to find when a similar failure shows up on another customer's. Supplier CAR Response Desk is designed to be that kind of record. It records and calculates; it is not advice or certification, and the analysis stays with your team.

If that is useful, join the waitlist to hear when the Response Desk opens.

Put it to work

Compare plans on the pricing page.

Get the next guide

Practical guides for supplier quality teams, delivered straight to your inbox.