Error Proofing: Poka-Yoke and Process FMEA

← All maturity guides

SEF Reference Guide G08  |  Manufacturing OpEx Maturity Assessment

What it is

Error proofing is about stopping defects before they exist. People make mistakes: they pick the wrong part, skip a step, or fit something backwards. Inspection finds some of those mistakes after the fact. Error proofing designs the process so the mistake cannot happen, or is caught the moment it does.

Two tools do the work. Poka-yoke (Japanese for “mistake proofing”) is the physical device or method: a locating pin, a sensor, a counter, a shaped fixture. Shigeo Shingo developed and named the approach at Toyota in the 1960s. Process FMEA (PFMEA) is the thinking tool: a structured team review of each process step that asks how it could fail, how bad that would be, how likely it is, and how well current controls would prevent or catch it. PFMEA tells you where to put your poka-yoke.

Used together, they turn error proofing from a reaction to the last complaint into a planned program. FMEA grew out of US military reliability work (MIL-P-1629, 1949), spread through aerospace and automotive, and today most manufacturers follow the AIAG and VDA FMEA Handbook (2019).

Why it matters

• Quality: A good poka-yoke gives near zero defects for that failure mode, every shift, without relying on attention.
• Cost: Preventing a defect at the station costs a fraction of catching it at final test, and far less than a field failure or recall.
• Safety: The same thinking prevents injuries, such as two-hand controls and interlocked guards.
• Speed of launch: A PFMEA done before start of production finds risks while they are still cheap to fix.
• Less inspection: Once a failure mode is truly error proofed, you can reduce the inspection effort for it.

Key terms

Term	Meaning
Poka-Yoke	Japanese for “mistake proofing.” A device or method that prevents an error from being made, or makes it immediately obvious so it cannot become a defect.
FMEA	Failure Mode and Effects Analysis. A team-based risk method that identifies how a product or process could fail, rates each failure mode for severity, occurrence, and detection, and drives actions to reduce the highest risks before they reach the customer. Process FMEA (PFMEA) applies it to manufacturing steps.
RPN	Risk Priority Number. Severity x Occurrence x Detection, each rated 1 to 10. Newer FMEA practice uses Action Priority (AP) tables instead of RPN alone.
Action Priority (AP)	The AIAG and VDA method that assigns High, Medium, or Low priority from the combination of S, O, and D ratings, with severity weighted first, then occurrence, then detection.
Failure mode	The way a process step fails to meet its requirement (for example, “bolt under-torqued”).
Effect	What the failure does to the next operation, the customer, or the end user.
Cause	Why the failure mode happens (for example, “wrong torque program selected”).
Severity (S)	Rating of how serious the effect is. Top ratings (9 and 10) cover safety and regulatory effects.
Occurrence (O)	Rating of how likely the cause is, given current prevention controls.
Detection (D)	Rating of how well current detection controls catch the failure before it leaves. A low number means reliable detection.
Prevention control	Something that stops the cause from happening, such as a fixture that only fits one way.
Detection control	Something that finds the failure after it happens, such as a vision check.
Control (shutdown) poka-yoke	Stops the process or makes the error physically impossible.
Warning poka-yoke	Alerts the operator with a light, sound, or signal but does not stop the process.
Contact, fixed-value, motion-step	Shingo’s three detection methods: check a physical feature, check a count, or check that steps happened in order.

How to do it

Poka-yoke

1. Start from real defects. Pull the top defect and complaint list. For each, trace the mistake that caused it.
2. Choose the strongest fix available. In order of preference: eliminate the step, make the error impossible (control), detect and stop, then detect and warn.
3. Pick the method. Contact (pins, guides, shapes, sensors that check a feature), fixed-value (counters, kitted trays with one slot per part), or motion-step (light curtains, pick-to-light, interlocks that enforce the sequence).
4. Keep it simple. Many of the best devices cost little: an asymmetric fixture, a shadow board, an offset hole.
5. Verify it daily. Every poka-yoke needs a check at shift start, often with a known-bad “red rabbit” test part. Log the result. A device that has been bypassed is worse than none, because people trust it.
6. Register it. Keep a list of every device, the failure mode it covers, its check method, and its owner.

Process FMEA

1. Plan the scope. One product family or line. Form a cross-functional team: process engineer, quality, operators, maintenance, and design where possible.
2. Map the process. Use the process flow diagram. List every step and its requirements.
3. List failure modes, effects, and causes for each step.
4. Record current controls for prevention and detection.
5. Rate S, O, and D using your agreed rating tables. Assign Action Priority. If your customer still requires RPN, calculate it too, but do not use an RPN cut-off (such as “act above 100”) as the only rule. A severity 9 or 10 item needs attention whatever its RPN.
6. Act on High priority first. Prefer prevention and poka-yoke over added inspection. Name an owner and a date for each action.
7. Re-rate after actions and update the control plan (see G10) so the new controls are part of daily work.
8. Keep it alive. Review the PFMEA whenever there is a process change, a new defect, or a customer complaint.

What each maturity level looks like

QS-3 Error Proofing (Poka-Yoke and FMEA)

Level	Assessment wording	What it looks like on the floor	How to move up
1	Error proofing is only applied after a defect or accident has occurred.	Devices appear only after a complaint or injury. Most checks rely on operator attention.	Train team leaders and engineers in poka-yoke and run a first pass on the top 10 defects.
2	Poka-yoke is understood and used in some places.	Some stations have fixtures, sensors, or counters. Coverage depends on individual engineers.	Build the poka-yoke register and apply devices to new processes before launch, not after complaints.
3	As Level 2, plus poka-yoke is applied proactively across the operation.	Devices are designed into new equipment and processes. Daily verification checks are logged.	Train a core team in PFMEA and complete one on the highest-risk product line.
4	As Level 3, plus Process FMEA is understood and used in some places.	PFMEAs exist for some lines and drive action lists. Ratings are team-based, not one engineer’s guess.	Extend PFMEA to all lines and link each High priority item to a poka-yoke or control plan entry.
5	As Level 4, plus a fully integrated program of poka-yoke and Process FMEA is in place.	PFMEA, control plan, and poka-yoke register are linked. Complaints trigger PFMEA updates.	Sustain: review PFMEAs on every change, track device verification, and feed lessons into new product launches.

Evidence to look for

Document

• Poka-yoke register with failure mode covered, check method, and owner.
• Daily poka-yoke verification logs.
• PFMEAs with dates, team members, ratings, actions, and re-ratings.
• Links between PFMEA, process flow, and control plan.
• Rating tables in use (AIAG-VDA or company-specific).

Observation

• Are devices in place and working? Try the red rabbit test with the team leader.
• Look for bypassed sensors, taped-over switches, or disabled interlocks.
• Do operators know what each device protects against?

Statement (questions to ask and who to ask)

• Operators: “What mistakes can happen at this station? What stops them?”
• Process engineers: “Show me the PFMEA for this line. What was the last change? What High priority items are still open?”
• Quality manager: “When a complaint comes in, how does the PFMEA change?”

Management report

• Number of High priority PFMEA items open and closed.
• Poka-yoke verification compliance (checks done versus due).
• Defects and complaints for failure modes rated as controlled in the PFMEA (these point to ratings that were too optimistic).
• Share of new processes launched with a completed PFMEA.

Common pitfalls

• FMEA as paperwork. One engineer fills in the sheet for an audit. Without the team, the ratings are guesses and the actions never happen.
• Chasing RPN thresholds. Teams nudge ratings to get under a cut-off. A high severity risk stays high risk no matter how the numbers multiply.
• Adding inspection instead of prevention. Another visual check lowers the detection rating on paper but does little in practice.
• Unverified devices. Sensors drift and fixtures wear. Without a daily check, a poka-yoke can fail silently for months.
• Warning devices ignored. A beep that sounds all day becomes background noise. Where the risk is high, stop the process.
• FMEA done once. A PFMEA that has not changed since launch no longer describes the process.

Template

PFMEA Worksheet Header and Columns

Field	Entry
Company / plant / line
Product family / part numbers
PFMEA number / revision / date
Team members and functions
Customer / special characteristics

Process step	Requirement	Failure mode	Effect	S	Cause	Prevention control	O	Detection control	D	AP (H/M/L)	RPN (if required)	Action	Owner	Due	New S / O / D / AP

Poka-Yoke Register

ID	Station	Failure mode prevented	Type (control / warning)	Method (contact / fixed-value / motion-step)	Verification method	Frequency	Owner	Linked PFMEA line

First 30 days

Week 1: List the top 10 defects and complaints from the last 12 months. For each, identify the human error behind it.

Week 2: Walk the floor and build the poka-yoke register. Add a daily verification check for every existing device.

Week 3: Brainstorm and install simple poka-yoke for at least three of the top 10. Train team leaders on control versus warning devices.

Week 4: Form a cross-functional team and start a PFMEA on the highest-risk line. Agree the rating tables and whether you will use Action Priority, RPN, or both.

Further reading

• AIAG and VDA, FMEA Handbook (2019): https://www.aiag.org/training-and-resources/manuals/details/FMEAAV-1
• ASQ, “What is FMEA?”: https://asq.org/quality-resources/fmea
• ASQ, “What is Poka-Yoke? Mistake and Error Proofing”: https://asq.org/quality-resources/mistake-proofing
• Shigeo Shingo, Zero Quality Control: Source Inspection and the Poka-Yoke System (Productivity Press, 1986).
• D. H. Stamatis, Failure Mode and Effect Analysis: FMEA from Theory to Execution, 2nd edition (ASQ Quality Press, 2003).