QUALITY IN THE FIRST PERSON
Failure of Inspection
The consequences of layering on quality checks
by Alan L. Austin
I am a quality prodigal. I don’t mean to say that I am very good at spending recklessly, but rather that I am like the young man from the parable1 who returns home after straying down other paths.
After many years in senior-level financial and administrative positions, I recently decided that it was time to come home to quality. I have been rereading many of the classics as part of my rehabilitation, especially the writings and teachings of W. Edwards Deming. Although I never met the legend, I did exchange correspondence with him and attribute my entering into the continuous improvement arena to his influence.
One of Deming’s 14 points for the transformation of management had to do with inspection as a means to achieving quality outcomes. He advised leaders to "cease dependence on inspection to achieve quality."2 Such an admonition strikes many as naïve, perhaps even fool-hardy. Those same individuals will say inspection has always been an effective means whereby we ensure that the customer receives the right thing. They will often wrap their response in the flag of pleasing the customer.
Deming’s response is that we should "eliminate the need for inspection on a mass basis by building quality into the product in the first place."3 Many organizations have applied the Deming philosophy to improve operational performance and profitability, yet many organizations continue to depend on inspection. Why the reluctance to reduce our dependence on inspection?
A useful definition of inspection comes from the Merriam-Webster Dictionary, which states that inspection is the "checking or testing of an individual (or thing) against established standards."4 In most organizations, inspection is used as a tool to make sure the right thing is going to the right customer under the right circumstances.
The challenge is that inspection often occurs at the end of the production line. As a result, the information gained from identifying problem pieces comes too late to make adjustments in the process, let alone isolate places in the process that are causing the problem. More inspection won’t help in those situations. In fact, more inspection will end up being more expensive and even less likely to improve the overall quality of the deliverable.
Too many checks
Let me give an example. When I first began my career in quality, I worked for a small company in Houston that would package, palletize and warehouse material for its customers. It would then ship the material to the customer when the customer needed that particular material in its manufacturing process. As a result, the company had many different materials in the warehouse that might look the same from the outside but were unique to the needs of a specific customer.
At about the time I went to work for the company, it started getting a number of complaints about the wrong material being shipped to customers. The existing process was that a forklift driver was given a pull ticket that would identify what palletized material needed to be pulled from what location of the warehouse and placed on a particular truck for shipment.
After the material was on the truck, a checker (inspector) would review the material and paperwork before approving the departure of the truck. Because many of the customers were overseas, the cost of the corrective action associated with incorrect shipments under the current process was prohibitive. The existing process of inspection was not stopping the incorrect shipments, so management decided that something needed to change.
The company’s initial solution was to assign a head checker to every group of checkers. These head checkers would verify the work of the checkers before a truck was allowed to leave the compound. Therefore, the new process was that a head checker would check the checker, who had verified the work of the forklift driver, who was responsible for checking his or her own work prior to the checker’s check.
Instead of improving the situation, the number of shipping errors actually increased. The company responded by requiring that the facility manager (FM) check the work of the head checker and sign off on the paperwork before the truck could leave. When that wasn’t sufficient to fix the problem, it was decided that the general manager (GM) or a vice president (VP) would need to check every load and sign off on its correctness prior to shipping.
That meant that in this newer process, a VP or GM must check the work of the FM, who verified the work of the head checker, who had checked the checker’s work after he or she had checked the work of the forklift driver, who was supposed to be checking his or her own work during process of loading the truck. At this point, the company was lucky to get anything out the door, let alone on time.
Being new to the organization, I felt that I needed to better understand the existing shipping process before proposing any changes. I recognized the cost of shipping errors was such that this process needed to be fixed as quickly as possible if the company was to stay in business. It was also clear to me that adding additional layers of inspection had not gotten the job done.
Getting to the root
To initiate my process improvement effort, I met with forklift drivers to map out the current process and to identify some of the reasons why shipping errors occur. We determined that a significant number of the errors happened on the night shift, so we collected additional data specific to nighttime operations.
A few factors seemed to be the major causes of shipping errors:
- Poor lighting in the warehouse.
- Faded paint identifying warehouse locations.
- Smeared packaging labels on the pallets.
- Several non-functioning head lamps on the forklifts.
The compounding effect of nighttime operations made the impact of these factors far greater than they were on the day shift. By addressing those issues through relatively simple changes, the number of shipping errors dropped significantly. We were also able to improve on-time deliveries over previous levels while removing costs associated with the additional inspection that had been bolted onto the process.
Building quality in
I experienced a more recent example of inspection’s potential futility while serving as a senior executive with an American-style university in West Africa. We would regularly host diplomats and dignitaries who would speak to our community regarding important and timely issues. Not long before I left Africa, we had the opportunity to hear from the Irish Ambassador to Nigeria, who would give a lecture titled, "The Peace Agreement in Northern Ireland—Lessons for Nigeria?"
When I arrived at the lecture hall, I was presented with a program that included a brief biography of the ambassador and an introduction to the lecture topic. This material had been prepared in the public relations department and reviewed by several individuals. A portion of one paragraph in particular jumped off the page at me. It read:
"Like Nigeria like Ireland, you would say, because the Irish republic situated on Europe’s third largest island and the Earth’s 20 largest has been the throes of violence."
This was not the first time poor quality documents had been published. The university had added additional reviewers over time in an effort to reduce these occurrences. Those who had reviewed the document were all educated, well-trained and competent professionals. The final review (inspection) was done by the president of the university, an American academic, who had inserted herself into the process in the hopes of eliminating the embarrassment associated with poor communications. Several people had reviewed the document, yet it contained meaningless and nonsensical text.
Clearly, inspection will not fix problems associated with a faulty process, but why does additional inspection often make the problem worse? The reason is that if I know someone else will be inspecting my work, I may count on them to catch anything I miss. Unfortunately, they might be thinking the same thing, namely that because other people have already looked at this, it’s probably fine. As a result, the number of errors is likely to increase.
All the inspection in the world can’t save a process that has been designed to fail. Deming was right. We need to build quality into the process. It can’t be added later on.
- "The Parable of the Prodigal Son," Luke 15:11-32, Holy Bible, King James Version.
- W. Edwards Deming, The New Economics, MIT Press, 2000.
- "Inspection," Merriam-Webster Dictionary, 11th edition, 2008.
Alan L. Austin is the director of business development for Breakthrough Consulting Group in Provo, UT. He has an MBA from the University of Houston. Austin is an ASQ member and an ASQ-certified Six Sigma Black Belt and quality engineer.