How do the FMEA and control plan connect?

Through special characteristics and recommended actions. The FMEA identifies high-risk failure modes and recommends prevention or detection controls, and those controls become rows on the control plan with a spec, method, frequency, and reaction plan. Floor failure data then flows back to update the FMEA's occurrence and detection ratings, closing the loop.

What is a special characteristic?

A special characteristic is a product or process feature whose variation could meaningfully affect safety, regulatory compliance, fit, function, or customer satisfaction. It is identified from the design FMEA, process FMEA, and customer drawings, and once flagged it must be carried on the control plan with tighter control, forming the audit trail that links the two documents.

Can one document replace the other?

No. They answer different questions and are used by different people at different times. Collapsing them into one spreadsheet feels efficient but quietly destroys the risk ranking that should decide where inspection effort goes, and it leaves operators without a clear instruction for what to check and how to react when a check fails.

How do you keep the FMEA and control plan in sync?

Assign one owner accountable for the linkage, trigger a paired review of both documents on any process change, cross-reference special characteristics by ID rather than memory, feed real floor and SPC data back against the FMEA ratings on a schedule, and audit the chain from drawing to FMEA to control plan rather than auditing each document in isolation.

Control Plan vs. FMEA: How They Work Together

Q: What is the difference between a control plan and an FMEA?

An FMEA is a risk analysis that lists how a process or product can fail and ranks each failure by severity, occurrence, and detection. A control plan is an operational document that names, for each important characteristic, the specification, measurement method, sample size, frequency, and reaction plan. The FMEA is where you analyze risk; the control plan is where you act on it.

Q: Which comes first, the FMEA or the control plan?

The FMEA comes first. You cannot write sensible controls until you know which failures matter, so the process FMEA is developed during process design and the control plan is drafted from its recommended actions. Writing the control plan first and back-filling an FMEA to justify it reverses the logic and defeats the purpose of ranking risk before spending on controls.

Q: Are the FMEA and control plan required?

In automotive, aerospace, and medical device work built to customer standards, both are typically contractually required deliverables, often alongside a full APQP package. Outside those industries the formality varies, but any operation with characteristics that matter benefits from separating risk analysis from the operational controls that respond to it.

The FMEA finds the risk; the control plan controls it. Two documents, one closed loop, connected by the special characteristic.

The FMEA and the control plan are two halves of one system: the FMEA finds the risk, and the control plan controls it. The FMEA asks "what could go wrong, how bad, and how likely?" The control plan answers "so here is exactly what we check, how, how often, and what we do when it fails." They are different documents with a hard dependency, not competitors.

People pit them against each other because both live in the same APQP binder and both list process steps and characteristics. But they answer different questions at different times, and confusing them is how organizations end up with a control plan full of inspections that guard against nothing and an FMEA full of risks nobody controls. This guide draws the line between them and shows the thread that has to run from one to the other.

What is the difference between a control plan and an FMEA?

An FMEA (Failure Mode and Effects Analysis) is a risk analysis: it lists the ways a process or product can fail, rates each failure for severity, occurrence, and detection, and ranks them so the worst risks get attention first. A control plan is a control document: for each characteristic that matters, it names the specification, the measurement method, the sample size and frequency, and the reaction plan when a check fails. The FMEA is analysis; the control plan is instruction.

Put simply: the FMEA is where you think and the control plan is where you act. The FMEA can recommend a control, but it does not tell an operator what to do at 6 a.m. on the line. The control plan does exactly that, and it exists because of the risks the FMEA found. Remove the FMEA and your control plan is a list of guesses; remove the control plan and your FMEA is a stack of worry with no teeth.

Dimension	FMEA	Control Plan
Question it answers	What could go wrong, how bad, how likely, how detectable?	What do we check, how, how often, and what do we do when it fails?
Type of document	Risk analysis (living, ranked)	Control / instruction (operational)
Key output	Prioritized failure modes and recommended actions	Characteristics, specs, methods, frequencies, reaction plans
Who mainly uses it	Cross-functional team; engineering	Operators, inspectors, line leaders
When it changes	New failure mode, field issue, or process change	When a control changes, or the FMEA recommends a new one
Owner of the link	Feeds special characteristics down	Must trace each control back up to a risk

Same process, two jobs: the FMEA prioritizes risk; the control plan operationalizes the response.

How do the FMEA and control plan work together?

The connection runs one direction and then loops back. During process development, the process FMEA identifies failure modes and their causes, and for the high-risk ones it recommends controls, either prevention (stop the cause) or detection (catch the failure). Those recommended controls become rows in the control plan: the characteristic to watch, the method, the frequency, and the reaction plan. The control plan is, in large part, the FMEA's recommended actions made concrete.

Then the loop closes. When the control plan runs on the floor and a characteristic keeps failing, that real occurrence data flows back into the FMEA: the occurrence rating was wrong, or the detection control is weaker than assumed. The FMEA gets updated, which may recommend a stronger control, which updates the control plan again. This is the closed loop AIAG's core-tools methodology is built around, and it is why both documents are called "living."

The FMEA recommends controls that populate the control plan; the floor's real failures flow back and re-rank the FMEA.

What is the thread that connects them?

The connective tissue is the special characteristic. A special characteristic is a product or process feature whose variation could meaningfully affect safety, regulatory compliance, fit, function, or customer satisfaction. It gets identified from the design FMEA, the process FMEA, and customer drawings, and once flagged, it is supposed to be carried on the control plan with a tighter control than an ordinary feature.

That is the audit trail an assessor walks: pick a special characteristic on the drawing, find it flagged in the FMEA with its failure mode, then find it on the control plan with a real control and reaction plan. If the chain breaks, if a critical characteristic is in the FMEA but missing from the control plan, or controlled on the plan but never risk-assessed, the system is not linked, it is two documents sitting near each other. Product characteristics on the control plan map to potential failure modes in the process FMEA, because a failure mode is simply that characteristic produced wrong.

An assessor walks this chain for one characteristic. A missing link, not a missing document, is the finding.

Where this linkage comes from

The FMEA-to-control-plan linkage is codified in the automotive core tools, where APQP, FMEA, and the control plan are designed to work as one system: the FMEA is a development deliverable, and its recommended actions and special characteristics are expected to flow into the control plan, which in turn feeds Statistical Process Control on the identified characteristics. The FMEA methodology itself was harmonized across the industry in the 2019 AIAG & VDA FMEA handbook.

Sources: AIAG, Automotive Core Tools (APQP, FMEA, Control Plan, SPC) · ASQ, Failure Mode and Effects Analysis (FMEA)

Which comes first, the FMEA or the control plan?

The FMEA comes first, because you cannot write intelligent controls before you know what you are guarding against. In an APQP timeline the process FMEA is developed during process design, and the control plan is drafted from it and finalized as the process is validated. Writing the control plan first, then back-filling an FMEA to "explain" it, is a common and backwards habit: you end up with an FMEA reverse-engineered to justify whatever inspections you were already doing, which defeats the point of ranking risk before spending on controls.

There is a practical exception. Organizations maintain three control plan phases, prototype, pre-launch, and production, and the FMEA is maturing in parallel across those phases. The two co-evolve. But the logical dependency never reverses: a control exists because a risk was identified, not the other way around. If you can point at a control-plan row and no one can name the failure mode it addresses, that row is ceremony.

How do you keep them in sync?

Drift between the two documents is the most common finding in a quality audit, and it is almost always caused by change happening in one place and not the other. Someone adds a poka-yoke on the line and never updates the FMEA's detection rating; someone loosens an inspection frequency and never checks whether the FMEA still justifies it. Keeping them synchronized is a discipline, and here is a workable one.

Assign one owner to the linkage. Not one owner for the FMEA and another for the control plan, one person accountable that every high-risk failure mode has a matching control and every control traces to a risk.
Trigger a paired review on any process change. A tooling change, a new supplier, a machine swap, or a new failure mode in the field all require opening both documents in the same session, never just one.
Cross-reference by characteristic, not by memory. Give each special characteristic an ID and carry it on both documents so an auditor, or a search, can jump from the FMEA row to the control-plan row instantly.
Feed floor failure data back on a schedule. Review actual reject and SPC data against the FMEA's occurrence and detection ratings; if reality disagrees with the rating, update the FMEA and re-examine the control.
Audit the chain, not the documents. Once a quarter, sample a few special characteristics and walk drawing to FMEA to control plan to reaction record. Broken chains are the real defect.

Do you always need both?

If you build to a customer standard in automotive, aerospace, or medical, yes, and they are contractually required deliverables, often alongside a full APQP package. Outside those industries the formality varies, but the logic does not: any operation with characteristics that matter benefits from separating "what could hurt us and how badly" (the FMEA's job) from "here is exactly what we do about it" (the control plan's job). Collapsing both into one spreadsheet feels efficient and quietly kills the risk ranking that should decide where your inspection money goes.

The failure everyone shares, regulated or not, is drift: the two documents fall out of step because change is captured on paper in one place and never propagates. When process changes, control results, and reaction records are captured digitally at the station, the FMEA and control plan can actually stay coupled, a loosened frequency or a new detection method is visible against the risk it was supposed to cover. That live coupling between risk and control, on top of your existing process with no rip-and-replace, is what Harmony's quality intelligence platform is built to keep honest. The same discipline underpins capability reporting, which is why control-plan characteristics show up again in Cp, Cpk, Pp, and Ppk studies, and why nonconforming output found by a control has to route through your control-of-nonconforming-product process. Even a well-run batch production line lives or dies on whether its control plan reflects the risks the FMEA actually found.