10.5 Error Prevention and Proofing

Improving Processes by Preventing Errors

You will often see news about errors in organizations leading to losses—and sometimes tragedies—occurring across distinctive industries. For example, many preventable process errors occur in hospitals across the U.S. One such tragedy is a neonatal intensive care unit’s healthcare staff accidentally harming babies by mistakenly administering adult medicine doses.

Such situations highlight the need to manage standardized work, which you will study later in this textbook. In the context of the Improve phase, we can anticipate that you need to detect the failures to follow standardized work proactively. A way to improve specific processes is to work with staff to make sure they can follow procedures and policies to improve their training.

Imagine that after running the previous DMAIC phases, you conclude that a worker has been cutting corners in his job. During the Analyze step, you discover that the process incorporates policies regarding how to perform the work, but the staff is not following them. Upon further analysis, you find that the worker has an intensive workload and has pressure to work faster than he should. In such a case, working with the staff can lead to process improvements that make procedures more realistic and prevent errors.

A vital way to improve a process is to ascertain quality at the source (jidoka in Japanese) by detecting and preventing errors. The crucial practice within the lean approach of avoiding mistakes goes back to the beginning. At Toyota, before the company had even begun to produce cars, Sakichi Toyoda designed a weaving loom that automatically stopped when the thread broke. This feature prevented the machine from producing items with defects.

The crucial point is that 100% inspection is not 100% effective. Preventing errors is better than post-inspection for many reasons. In the context of the present topic, first, if you rely on people to inspect the work, you introduce the risk of human error into the process; workers get tired, become distracted, or forget. Second, when multiple workers are responsible for inspecting for a defect, there is an all-too-human reaction of complacently assuming another worker will find the defect. Instead, error-proofing methods and standardized work (which will be discussed in a later topic) can improve processes.

Error-Proofing

Error-proofing consists of creating devices or methods that either prevent defects or inexpensively and automatically inspect the outcomes of a process each time to determine if the quality is acceptable or defective. You can think of error-proofing as a mindset and approach that calls for creativity among process participants. It is not a specific technology or device. With that brief intro, there are four main ways to error-proof a process. Let’s take a look at each of them.

Making the Error Impossible to Happen

We can likely agree that the best possible scenario would be when error-proofing 100% physically prevents an error from occurring.

Imagine you’re filling up the gas tank of your car, and you’re tired and grab the first nozzle you see, not bothering to read the signs. You grab the diesel nozzle. Diesel fuel in an unleaded gas vehicle can be harmful to the engine. To solve this problem—and to protect drivers from a dangerous error—the diesel nozzles are larger than unleaded nozzles; this means that the diesel nozzle won’t fit into an unleaded gas vehicle. The different nozzle sizes make it impossible for you to make the error of refueling your unleaded gas vehicle with diesel. Such a device is more effective than placing warning signs in or around the fuel pump.

Facing a similar problem, many hospital rooms have connections to different gases, such as oxygen, medical air, carbon dioxide, and nitrogen. Each has a specific function, and mixing them can lead to severe or fatal consequences. Have you ever noticed that their connectors in the wall are colored differently (see figure below)? However, hospital staff often find themselves in high-pressure, fast-moving emergencies; mistakes happen. Error-proofing the process by using different pins to plug the lines into the connector physically prevents those errors from occurring.

You can look at physically error-proofing steps in your process as an improvement solution to prevent errors and enhance output.

Making It Harder for the Error to Happen

As much as we would like to completely error-proof a process, it’s not always possible to do so. The next best alternative is to make it more challenging to incur the error. For example, think of the last time you worked in a Microsoft Word document. Did you try to close the file before you saved it? This is one of the most common potential errors, and Microsoft has done their best to make it harder to make this mistake by having a confirmation step in the form of a pop-up box to prevent or minimize data loss. See figure below.

Similarly, picking errors in a distribution facility is a significant barrier to productivity and profitability improvement. Picking is when workers choose individual items in the fulfillment facility to satisfy customers’ orders. Warehouse picking errors negatively impact customer satisfaction and can occur somewhat frequently due to the pace of work.

Life in a warehouse is often fast-paced and intense; associates can become fatigued over the course of their shift and thus less attentive. Adding collaborative mobile robots, such as the one in the figure below, can reduce picking mistakes by leading associates to the correct pick location and displaying order items and quantities required to satisfy a particular customer order. In general, making the error harder to happen can be an improvement solution to certain process problems.

Making the Error Occurrence Obvious

Consider shipping packages: many boxes contain fragile goods, so rough handling can damage or break the cargo content of packages. Many packages nowadays travel thousands of miles and change transportation modes many times during their journey. The possibility of mishandling the package and damaging the content is high. Shock indicators are tamperproof labels that you can place outside the package before transportation.

Shock labels indicate when fragile products have been exposed to potentially damaging impact during transit or storage. They are mechanically activated and turn bright red when impact occurs, alerting customers about mishandling. The devices do not prevent the error from happening, but they make it evident that the error took place. If you cannot prevent error from happening, making it evident can be an improvement solution in some situations.

Making the System Robust to the Error

Finally, another way to improve a process by proofing it is to make the operation sturdy so that it can tolerate the error. A classic, everyday life example is the possibility of driving away from the gas station with the gas nozzle still attached to the car. Such an error could lead to spills and a possible explosion. While the error could not be prevented, the system is designed so that if a driver drives off, the pump’s quick-release valve snaps away and cuts off the flow of gasoline, protecting against the potential consequences.

Another example of this can be found in an aviation setting. A tail strike occurs when the tail of an airplane contacts the runway during takeoff or landing. It can cause substantial damage to the aircraft’s fuselage, consume time and resources to repair, and even be catastrophic. Although environmental factors such as wind gusts can cause it, tail strikes often occur due to human error, such as rotating the nose too rapidly during takeoff or raising the nose too sharply during landing. Anticipating the issue, aircraft manufacturers reinforce the tail and carry out flight tests during the certification process.