Would you wear a blindfold to drive to the grocery store? Maybe it’s a journey you’ve taken hundreds of times, and it’s only a quarter of a mile down the road. You could probably drive it in your sleep and describe in detail the turns, crossings, and traffic lights that stand in the way of your destination.
Of course, you wouldn’t drive blindfolded. But that’s essentially what manufacturers are doing with their supply chain’s goods without acceptance sampling —they’re crossing fingers and hoping vendor products meet requirements.
But not all companies use acceptance sampling. The scary truth is that some don’t even inspect incoming vendor products at all. Others rely solely on Certificates of Analysis (COAs) that they receive from vendors. Whether using COAs or by not inspecting (and verifying quality levels) at all, they are placing a great amount of trust in their vendors. You know the saying, “trust, but verify.” By not verifying with acceptance sampling, it’s like organizations are driving their manufacturing processes blindfolded, hoping everything will be okay. No matter how well-intentioned, hope is not a strategy.
Acceptance sampling is the answer. It helps manufacturers minimize inspection costs, manage risk, and prevent off-quality product from entering their production processes.
ProFicient™, InfinityQS’ quality management software system, is designed to help manufacturers achieve quality excellence. It was designed to help organizations transform their operations with statistical process control (SPC) methods. ProFicient also includes expansive acceptance sampling capabilities. No other real-time quality software on the market today supports both extensive SPC and acceptance sampling features in the same application.
And if you work with a supply chain of any kind or size, using both SPC and acceptance sampling can prove to be a game-changer. Allow me to explain why acceptance sampling is so important and how InfinityQS can help.
What is Acceptance Sampling?
According to my colleague Jude Holmes, InfinityQS Client Solution Engineer, in the blog
Exploring Acceptance Sampling, acceptance sampling was created to refine manufacturing and reduce unacceptable lots. “A sample is picked at random from the lot and, based on information that was yielded by the sample, a decision is made regarding the disposition of the entire lot. In general, the decision is either to accept or reject the lot.”
Acceptance sampling occupies the middle ground between no inspection and 100% inspection. Because of the simplicity of its results, acceptance sampling has been given a bad name. It was criticized over the years as “just another set of inspection tools.” But that’s not true. Together with SPC, acceptance sampling can be used as part of a formidable quality improvement strategy.
Accept or Reject?
By itself, acceptance sampling is a gatekeeper, of sorts. As mentioned, in simplest terms, it is an accept/reject moment in manufacturing. But ProFicient quality management software makes it much more than that. Our software makes acceptance sampling an ally in the war against bad quality products and less-than-optimal supply chain quality.
Pass-Fail
Problems with quality typically begin with vendor materials that make up the foundation of your products. Let’s suppose you’re in the business of making aircraft or some other large, complicated vehicle. Manufacturing large metal
anythings requires a lot of rivets. Thousands upon thousands. Maybe millions, depending on what type (and size) of vehicle it is.
The person who must inspect received rivets has it tough. They’re certainly not going to check every single one. Can you imagine measuring and checking every one of 500,000 rivets in a single shipment? No way. Instead, out of a half million received rivets, they might check 50 as defined by their acceptance sampling plan.
Basically, the inspector needs to know—based on their chosen sample—if the lot passes their company’s stringent requirements. It could be a half a dozen or more different tests for each rivet. Acceptance sampling says that when a specific number of their sample fails, the collective lot fails. If the number of failures is less than the “reject” number defined by the acceptance sampling plan, then the lot is accepted. Generally, this is where most companies stop. Pass-fail. Yes-no. Accept or reject. Most companies don’t even save the data that allowed them to make critically important accept/reject decisions. That’s a mistake.
As an industrial statistician, I never understood why someone would ignore data. Yet, with acceptance sampling, that is exactly what happens. Once the lot has either been accepted or rejected, the data that lead to that decision is ignored or eliminated—as if that data had no other purpose or use than simply concluding whether or not to accept a lot of goods. What a waste!
In my 30+ year quality career as a statistician, I can plainly state that nearly every data set I have ever encountered includes valuable information. The same is true for the data gathered during acceptance sampling activities. So, don’t ignore that data. Instead, use it to further improve your quality, and the quality of your vendor’s products.
Take it a Step Further
Some of the manufacturers of larger vehicles (like aircraft) and their ilk use so many rivets, cut so many holes, and use so many parts that they couldn’t possibly source rivets from
one vendor. They might use five different vendors for rivets alone. So, why not look at the data and determine which vendor is serving them best?
The data have already been gathered using acceptance sampling techniques. That data can be analyzed again using the statistical tools that are already included with InfinityQS quality management software, ProFicient.
Using acceptance sampling data, you can:
- Determine which vendor (out of many) is providing the highest quality products
- Analyze how vendor quality changes over time
- Determine if vendor quality is improving or degrading over time
- Help vendors improve their products
- Identify which vendors should be rewarded with longer-term contracts
Acceptance sampling data should not be ignored or eliminated after the accept/reject decision has been made. Instead, that data should be used to continue to improve quality levels across the value chain.
Acceptance Sampling Reports
Beyond the ability to analyze acceptance sampling data using real-time statistical tools, ProFicient also creates reports that are specific to acceptance sampling.
ProFicient’s acceptance sampling reports use color coding to indicate acceptance (green) or rejection (red), inconclusive results (yellow), and lots that require retesting.
ProFicient even goes the extra mile to send email notifications to the appropriate parties when lots fail. That’s the kind of communication that keeps your entire team on the same page…and vendors’ quality in mind.
Use the Power of the Cloud
ProFicient’s sister product, ProFicient
™ on Demand (PoD), uses the power of the cloud to add even more analyzing capability to an already robust tool. PoD offers companies a way to analyze across all plants, all regions, and even all vendors—a single, unified data repository.
SaaS
PoD is a Software-as-a-Service (SaaS) offering. As I explain in another blog,
Shifting Manufacturing Quality Paradigms, “SaaS provides manufacturers with the capability to have
data collection and
visibility across the entirety of an enterprise. SaaS provides our quality improvement experts at the corporate level (from anywhere around the globe) to be able to sort, slice, and dice data anyway they want from across plants, across departments, across the enterprise, across regions. It just doesn't matter anymore; there are no limitations.”
The unified data repository offers manufacturers a way to keep all their data in one place, a place from which they can compare and contrast any value, line, or sample they wish. Seeing which vendor supplies the materials that generate the least amount of rejections just got a whole lot easier. That’s supply chain management in action.
So, don’t drive blindfolded. Be sure to use acceptance sampling to monitor the materials your supply chain sends you to make your products. And use acceptance sampling as more than just an accept/reject moment in manufacturing—analyze it to ensure that you’re always using the best materials (and vendors) possible to make your products.
Learn more about InfinityQS
ProFicient and
ProFicient on Demand.