Error Proofing
-
Speaker : WILLIAM LEVINSON
-
When : Tuesday, October 14, 2025
-
Time : 01 : 00 PM EST
-
Add To Calendar
Refer a Friend
William A. Levinson, P.E., FASQ, CFPIM is the principal of Levinson
Productivity Systems, P.C. and the author of several books on quality,
productivity, and management.
- Know what error proofing is
- Know why error proofing is important (and mandatory, where applicable, for IATF 16949:2016 users)
- Learn the relationship between error proofing and prevention and detection controls in failure mode effects analysis (FMEA)
- Learn the definitions of successive check inspections, source inspections, and self-check inspections. These are often 100 percent and ideally automated, as opposed to inspections by people
- Learn error-proofing techniques, as illustrated by case studies
Areas Covered
- Army Technical Publication (2014), ATP 5-19, Risk Management, makes it clear that engineering or technical controls are superior to administrative or behavioral controls, the kind that rely on vigilance and compliance. When a Shigeo Shingo case study begins with the proposition that mistakes are avoided through worker vigilance, it's a sure sign that mistakes were happening simply because it is possible to do a job wrong. If we do it enough times, we will eventually do it wrong.
- Error proofing embodies the "Can't rather than don't" principle from the Ford Motor Company, where physical controls made it impossible to do the job wrong.
- "Don't take it, don't make it, don't pass it along," where "it" refers to poor quality. (Cutcher-Gershenfeld, Joel, and Dan Brooks, Martin Mulloy. Inside the Ford-UAW Transformation. MIT Press, 2015.)
- Successive check inspections of work coming into a process support "Don't take it." These are detection controls that, while they don't keep the nonconformances from happening, do not allow them to waste the receiving process' resources and alert the supplier process of their existence.
- Source inspections during the process support the statement "Don't make it." These are prevention controls.
- Self-check inspections before the work leaves the process support "Don't pass it along." These are detection controls that keep nonconformances from getting to the following process and
alert the current process with audible and/or visual controls. - In all cases, these inspections are (ideally) automated to not require human labor or rely on human judgment.
- Error-proofing devices and procedures are Detection and Prevention controls as depicted in the AIAG/VDA (2019) manual on failure mode effects analysis (FMEA).
- Detection controls intercept nonconformances after they happen, but hopefully before they reach the next internal or external customer. They should also alert the production workers that something has gone wrong so they can correct it and stop the process to avoid the generation of more nonconformances.
- Prevention controls prevent the generation of nonconforming work in the first place.
- In all cases, engineering or technical controls that do not rely on compliance or vigilance are superior to administrative or behavioral controls that do. The former is capable of 100 percent inspection, such as automated inspections that ensure 100 percent quality in contrast to acceptance sampling plans that, by necessity, specify acceptable levels of poor quality (the acceptable quality level or AQL).
- Case studies illustrate error-proofing controls and their effectiveness as rated in FMEA. For example, an automated detection control that will detect the problem in-station, where it is generated, has a rating of 3 (High) on a 1 to 10 scale, with one being best and 10 being worst.
Who Should Attend
Quality managers, engineers, technicians, and others with responsibility for advanced quality planning (AQP) or PPAP
Why Should You Attend
Many Shigeo Shingo case studies begin with scenarios in which mistakes were to be prevented with worker vigilance; this hints that mistakes took place and poor quality was generated. The reason is not that workers aren't vigilant but rather because if it is possible to do a task the wrong way, it is only a matter of time before this happens. This is why engineering controls, or physical controls that prevent mistakes are superior to administrative controls that rely on compliance, attention, and vigilance.
To put this in perspective, the Automotive Industry Action Group's and German Association of the Automotive Industry's 2019 manual for failure mode effects analysis (FMEA) rates occurrence and detection controls on a 1 to 10 scale, with one being best and 10 being worst. Prevention controls prevent nonconformances from happening in the first place, and detection controls prevent them from escaping to the following internal or external customers. The only way to get a 1 Occurrence rating is with technical controls that make it impossible for the failure cause (previously known as the failure mechanism) to produce the failure effect. Administrative controls that rely on inspections by people cannot meanwhile earn better than a 6 Detection rating even if they "will" detect the failure mode or failure cause.
Machine-based detection methods are required for better Detection ratings.
The principle was known as "Can't rather than don't" at the Ford Motor Company, where, for example, workers were told, "Don't put your finger into the sewing machine when it is running." This resulted in 3 to 4 dangerous puncture wounds (as the tetanus vaccine had yet to be invented) every day. Installing a simple guard made it impossible to do so, and the injury rate dropped to zero. This is from a safety perspective, but the same principle applies to product realization. Shigeo Shingo's Zero Quality Control: Source Inspection and the Poka-Yoke System provides numerous examples.
-
$160.00
-
