How Statistical Process Control Was Misrepresented to US Audiences

Executive Summary

• Statistical process control was one leg of the methods applied by Deming and the Toyota Production System.
• While the other methods have been de-emphasized, statistical process control has been overemphasized, which is for a premeditated reason.

Introduction

Toyota is often put forward as an exemplar of what to follow — and there is a cottage industry in consulting which often explains what Toyota did inaccurately.

Most of the books on Toyota and the Toyota Production System (TPS) were written by consultants who have consulting services to sell. Therefore, they will often alter the actual history of what sets Toyota apart to appeal to executive audiences.

Who Controls Quality?

An essential part of the Toyota Production System is that responsibility for quality is distributed across the line employees that do the work. Statistical Process Control (SPC) is typically performed at the end of the process, picked up by US authors, and popularized to a significant degree by US manufacturers. But there is an essential and undiscussed reason why Statistical Process Control received so much emphasis when the Toyota Production System was communicated to the US audience.

The Overstatement of Statistical Process Control as Part of the Toyota Production System

Statistical Process Control was taught to Japanese manufacturers by Edward Deming and others that followed him.

However, the Japanese or Toyota Production System approach also starts with the quality orientation of empowered workers before the final output is produced. Statistical Process Control is the end of the quality chain. Statistical Process Control is an important part of the Toyota Production System, but Statistical Process Control is in no way the “center” of it. The other factors, far less emphasized by most US consultants, are not any less critical than Statistical Process Control.

Also, with hundreds of thousands of professional statisticians living in the US, and with statistics widely taught in the US when Deming was writing his books and consulting, it is confusing why Statistical Process Control was such an enormous revelation.

Statistical Process Control is the review of the number of out of specification items as part of a “sample” and then calculate the sample’s quality level. Statistical Process Control was designed to determine if this sample, which was intended to represent the production population, was high enough or if changes needed to be made to the manufacturing process. This is foundational statistics.

The Labor Input to Quality and Censorship by US Authors on the Toyota Production System

While Statistical Process Control was highly emphasized in books and in consulting to US audiences, specific areas were de-emphasized.

These Quality/TPS author/consultants learned to hide those things from their target audience, which were critical to implementing the Toyota Production System but may have lead to cognitive dissonance with what the audience would like to have believed. For example, Toyota gave its line production workers the ability to shut down the line if they thought production would lead to the quality problem. These workers were both empowered, and they also happened to be unionized, and they were well paid.

As of now, “labor unions” are a dirty word in the US, and it is critical to adjust the message so that nothing is said positively about them.

The authors knew this and that the executives were not going to want to empower the line workers in any way. Telling US executives the truth would have been a good way not to obtain a consulting deal.

Therefore, this critical aspect of the Toyota Production System was removed or self-censored from most of the literature on Toyota manufacturing. The original books written by Japanese authors like Taiichi Ohno and books by Edward Deming include essential aspects of the Toyota Production System that later writers removed.  This would be like providing someone with a recipe and leaving out several of the ingredients.

Perhaps unsurprisingly, the US manufacturers never attained the same type of benefits that the Japanese manufacturers had achieved from the Toyota Production System and related approaches imported from Asia, even though these programs were expensive investments on US manufacturers. It helped to spawn Six Sigma and several quality programs, which also tended to fall short of the originally intended objectives.

Cultural Differences?

The discrepancy was later chalked up to “cultural differences.” Japan was often described as a special place with a unique culture, where some things that worked in Japan could not be exported. This all seemed to make sense until Toyota began to build cars in the US. The quality levels attained did not match the lower level of the US manufacturers, but instead matched the quality of the vehicles that Toyota made in Japan!

At no time was it a point of discussion that perhaps the US consultants that received so many Japanese/Toyota-inspired manufacturing improvement projects accurately represented these systems to US manufacturers.

Statistical Quality Control as Trend

One hears far less about Statistical Process Control specifically and statistical quality control generally.

Statistical quality control was so highly studied in the 1980s. Soon after the statistical quality control trend, where we were told ad nauseous how important statistical quality control was, US companies began outsourcing manufacturing to low-cost countries. Countries like China have all different levels of quality in manufacturing to choose from, as his quote attests.

“From my experience with my clients and suppliers I’ve met in China, I can definitely tell that the importers are also responsible than the suppliers.

Sure, the suppliers can change the raw materials of the products, don’t pay attention to the manufacturing process and won’t even proceed correctly to a quality check or just skip it and so on.

But at the same time, a lot of importers come to China to produce “cheap” goods and still expect a good quality when they chose suppliers who are pretty famous for the low quality of their goods but they are extremely cheap so yes, they give them their production and surprise! The products won’t last more than 2–3 months.

You can produce high quality products in China depending on how you chose your supplier (make an audit for that) and if you make rigorous quality control inspections (at the minimum one).” – Sarah Bonaca, Quality Control Advisor in Asia

So many US companies had the choice and chose lower prices rather than quality. It is very well understood that the Chinese items that have replaced US manufactured items are, on average, considerably lower in quality. This brings up the question as to who serious.

This brings up the question as to who serious was about statistical quality control. Within just a few decades, US companies moved from being focused on statistical quality control and stating that Six Sigma quality was critical to accepting far lower quality from overseas manufacturing to be “competitive.” Strange. We can’t get companies to agree that overseas workers in manufacturing facilities should not work in slave-like conditions. What a setback for quality in all the dimensions.

Conclusion

Toyota and the Toyota Production System are often cherry-picked by people who want to emphasize an attribute or way of doing something they favor. And the presentation of the role of Statistical Process Control in the manufacturing success of Toyota is one example of this.

So the next time you hear about how important it is to be flexible in manufacturing. When Toyota’s name is used to support flexible manufacturing as a virtue, you can send that person this article for some more perspective.

References

https://www.quora.com/Why-are-Chinese-products-of-such-low-quality