Poka-yoke: preventing errors at work

Poka-yoke aims to prevent errors or make them apparent before they cause consequences. Safety applications require validation of the protective function.

In brief

A poka-yoke is a solution for preventing or detecting errors. A device intended for quality control must not automatically be accepted as protection for people.

Contents
  1. What poka-yoke means
  2. Prevention and detection are different
  3. Its relationship with machinery safety
  4. Hypothetical example
  5. How to check that it adds value
  6. Related concepts
  7. References

A–Z dictionary →

What poka-yoke means

Poka-yoke describes methods that help prevent execution errors or detect them in time. It may be a physical shape that prevents a part being fitted backwards, a presence check or a signal that makes an omission visible.

The concept is widely used in quality and process improvement. For prevention, its value is that it addresses task design rather than relying entirely on a person remembering every instruction on every occasion.

Prevention and detection are different

A prevention solution stops an incorrect combination or movement from occurring. A detection solution indicates that an error has already occurred so it can be corrected before continuing. This distinction changes the response needed.

An alarm may go unnoticed, be confused with others or arrive too late. The visibility of the signal, the time available and the expected action therefore need analysis, along with the conditions under which the device can fail.

Its relationship with machinery safety

Not every poka-yoke is a safety component. A sensor used to check assembly quality does not, by itself, establish that a machine is safe for someone entering its danger zone.

When the function prevents harm, its design and validation must meet the risk analysis and requirements applicable to that function. Guards, interlocks and isolation procedures remain necessary. Bypassing a guard because another warning exists to prevent product defects is not appropriate.

Hypothetical example

Two containers for different uses have very similar connections, creating a risk of interchange. A physically incompatible connection is considered, together with clear identification and a verification procedure. The solution reduces the possibility of connecting the wrong container.

Before implementation, strength, material compatibility, maintenance and the consequences of improvised adaptation are checked. The example illustrates a design principle, not a universal recommendation to modify equipment or installation connections.

How to check that it adds value

The review must include normal tasks, changeovers, cleaning, repair and foreseeable time-pressure situations. Listening to staff also helps: a device that is awkward or blocks correct operations can encourage unintended uses.

Record the error to be prevented, how a failure is detected and who may intervene. Then verify the result without creating dangerous exposure. The aim is to reduce reliance on human memory while maintaining reliable engineering controls, rather than blaming every subsequent incident on the system user.

Explore this approach further in Machinery safety.

References

  1. Lean Enterprise Institute: Error-Proofing. Official source
  2. EPA: Lean Thinking and Methods, TPM. Official source
  3. Government of Tasmania: Lean Principles. Official source

Editorial information

Publication date: 1 November 2026.

Editorial responsibility: Sabentis Editorial Team.

Author: Pablo RodríguezLinkedIn

Executive Vice President of the ORP International Foundation and Chief Financial Officer of Sabentis.

Request a Demo

Discover all that Sabentis can do for your organization.

Try Sabentis

request a demo
stars 5
GetApp Software Advice Capterra