Have you ever wondered why it is so easy for improved processes to go back to how they were performing before improvement? Why it always seems that, even if everybody follows the new instructions, the process still produces defects? Why it’s so hard to maintain the benefits in the process after the projects are done? If you have ever asked yourself any of these questions, you might be surprised to learn that the answer lies in the effectiveness of implemented process controls. The better and more robust your new controls are, the greater the resistance of the process to the gradual degradation that commonly occurs.
Regardless of the different types of process controls that are commonly put in place during a process improvement project (SOPs, visual management, kanbans, etc.), we have found that the most effective process controls are derived from the approach known as “poka yoke,” or “mistake proofing.” Processes that are mistake-proof are systems that are intolerant to defects, ideally outright preventing the occurrence of errors or, in the least, detecting them when they happen very early in the process.
Mistake-proofing devices are simple process/product design changes that make defect prevention/correction infallible and effortless. When combined with lean, Six Sigma or other process improvement methodologies, mistake-proofing can be a significant contributor to the maintenance of improved processes.
Characteristics of Mistake Proofing
Consider this example: As a naval officer, Fred was constantly amazed by the records that he had to review each day. Out-of-tolerance readings on these equipment logs were to be circled in red ink, and the log was to be annotated and signed in black ink only. These simple instructions were seldom followed, and the records continued to be written in every imaginable color. Fred was frustrated. No matter what he said, nothing seemed to make a difference. One day, one of Fred’s chief petty officers heard him grumbling about the issue, and then promptly went and collected all the pens, except black ones, from all the desks where the records were maintained. The chief then gave Fred the only red pen in the department. Problem solved!
The mistake-proofing concept is not new; it’s been around for decades. It has additionally gained popularity in recent years. According to its definition, mistake-proofing devices should meet three criteria:
When designing mistake-proofing solutions, it is always good to check if you are fulfilling those characteristics. The following list of questions and examples can offer you a guideline as you begin to think about how to mistake proof your processes:
- Is it always present/activated? In Fred’s story, there simply was not an alternative to using the black pens -- they were the only option available. So the mistake-proofing solution that the chief employed was always present, in all circumstances. Or you might think of a fixture that prevents items from being mounted incorrectly, like the notch on the old 3.5-inch disk drives. They would only fit into the slot on the computer one way.
- Is it effective? How many times is the error/defect not captured? For example, at convenience stores, people often unintentionally walk off with the pens that the checkout clerk needs for people to sign their credit card receipts. The simple solution of taping a large feather to the pen is effective mistake proofing that makes it very difficult for a customer to forget they have the pen.
- Is there an immediate action when an anomaly is detected? Preventing gasoline overflow when filling your tank is a mistake-proofed process now because of auto-shutoff mechanisms that have been nearly universally installed. This is an immediate action that takes place when a particular condition is being met.
- Does it add extra steps? Mistake proofing a process should not add complexity. In fact, in the best of circumstances, they make being successful in the new process simpler, aiding in compliance. The black pens, the feather on the pen in the convenience store, the auto-shutoff -- each one of these mistake-proofing solutions added no complexity to the process, and in the auto-shutoff example, the solution actually made it easier for the customer, too.
If your mistake-proofing devices violate some of these characteristics, they should be considered as underperforming. In some cases, underperforming mistake proofing can result in false alarms or missed defects which in turn will result in undermining the confidence in the whole improvement process. Failed process control on an improved process is still a failed improvement project, regardless of the improvement potential.
Designing a Mistake-Proofing Solution
While mistake proofing has a rich history in the manufacturing world, often people fail to see the implication for transactional or service environments until shown sufficient examples. Regardless of your opinion as to the use of them, red-light cameras are a form of mistake proofing (error detection, always on). The autocorrect function and grammar checking in word processing software or the autocomplete functions in search engines are other transactional examples.
But where do you start when you are looking at a process and are seeking to figure out how to insert mistake proofing into your process improvement project? At BMGI, we would approach a mistake-proofing problem with our innovation methodology (D4: Define, Discover, Develop, Demonstrate) in order to design and deploy a new mistake-proofed solution.
1. Define Mistake-Proofing Areas/Opportunities.
Depending on the nature of the process being improved, a process map or an illustrated parts breakdown might help to identify high-risk areas for mistakes. Alternatively, if the process already has a product, process or design failure modes and effects analysis (FMEA), then the high-risk areas have already been identified formally through the use of risk priority numbers (RPN). Regardless, the purpose of this stage is to identify and prioritize the opportunities for mistake proofing.
2. Discover Potential Solutions.
Once the critical few have been identified, then it is time to generate a list of mistake-proofing ideas for each opportunity. We recommend that groups generate at least five mistake-proofing ideas for each opportunity. The use of a mistake-proofing matrix, or principles of problem solving from TRIZ, can be used to brainstorm ideas based on the conditions of the processes/systems and types of errors/defects you want avoid.
3. Develop Potential Solutions.
Tools such as resource optimization and separation principles will help screen some of the potential solutions and ensure that the ones continuing to the final phase of the design process are effective, simple and inexpensive.
4. Demonstrate the Solution.
After narrowing down the potential solutions, we recommend piloting and/or prototyping two or three mistake-proofing ideas and then evaluating them to select the one to be implemented. The mistake-proofing characteristics covered above can be used as the evaluation factors for selecting the best solution.
Benefits of Mistake Proofing
There are numerous benefits that can be realized by focusing on mistake proofing as a core component of your approach to process control. In general, mistake-proof processes need fewer inspections, produce less scrap, and subsequently have reduced rework and repair investment. In the best of circumstances, you could produce a solution that creates zero errors/defects. As a consequence, mistake-proofed processes experience an increase in productivity, as the non-value added steps like inspections can be removed. Additionally, mistake-proof processes have less in-process inventory.
Of course, the business will also experience a reduction in cost of poor quality (COPQ). Although it is always difficult to estimate the full amount of COPQ, the cost of detecting and correcting an error internally is significantly less than recalling defective products or apologizing for incorrect services to customers.
Mistake proofing is a simple concept that can make a significant difference in the output of any process. When either creating a new process or improving an existing one, mistake-proofing solutions should be included as part of the result.
Mark McMurray and Jorge Garcia are senior client partners with BMGI. Both Lean Six Sigma master black belts, they bring years of experience to their roles in leading transformation efforts. Mark holds a master of science in operations management from Kettering University and a bachelor of science from the U.S. Naval Academy. Jorge has a degree in industrial and systems engineering from the Monterrey Institute of Technology in Monterrey, Mexico. He was also a naval pilot for the Merchant Marine School of Mexico.