Another month has gone by and everyone is getting settled in to work. It’s a long stretch until the next natural break (spring break for most) and coupled with the dreary weather, it makes the upcoming couple of months seem tiresome. It’s as good a time as any to lighten up the mood with another engineering discussion that has no immediate consequence, but hopefully sparks some internal conversation.
I’m going to go back to the list at the end with suggested topics and pick one for the newsletter this month. My random number generator has determined that this month’s topic will be commercial grade dedication. On the surface that is not very interesting, but from the perspective of what really matters in the process, I think the conversation could be much more engaging. At least that is my hope.
The new year has brought a renewed focus on advancing my business. I spent the last month putting together a long-term vision and focusing on the next steps. Part of that is realizing I am not an island and need to engage more with the engineering community. In that vein, I’m increasing my participation in the local ASME chapter and going to the Group Leadership Development Conference this year. If you’re going to, let me know!
Let us clear the air and disclose my relationship to this topic. Coming from an engineering consulting firm firmly planted in the nuclear industry, my perspective is related to commercial grade dedication as it is described in the ASME NQA-1 standard. I’ve spent a lot of time with that formulation but come to my own generalized view on what it all really means. It’s this generalized view that I’m going to talk about, but I realize that some of my terminology may be very industry specific, so please excuse me if I get wrapped up and forget to define something.
Commercial Grade Dedication (CGD) is a process of certifying (or dedicating) that products developed outside of a specific quality program (commercial grade item) will still meet the requirements of a product that had been developed under such a program within some constraints of use. Usually this is done by focusing on the design intent for a product and then ensuring, with an appropriate level of rigor, that the commercial item will be able to meet this design intent.
Note that we are “dedicating” some item to perform a task that was designed to have some level of quality control. Technically all items have some level of quality control, or we would pour water into our engines or shod our cars with wooden tires. We are always doing CGD, but this topic is more about how it relates to engineering specifications.
In its most simple form, CGD has four basic pieces:
We’ll take each one of these apart considering the simplest of all cases – replacing the batteries in your TV remote. In later sections I’ll break down each in a more generalized manner.
You’re blithely watching your favorite show but want to turn up the volume for that awesome car chase scene. Nothing happens. Because you’re a star engineering sleuth, you reason that the most likely culprit is depleted batteries (especially since you can’t remember the last time you replaced them). You need new batteries.
The function of the batteries is pretty obvious – it’s to provide power (probably around 3V and a few milliamps) to your remote control. That’s fully describes the function.
The critical characteristics are the attributes of the item to meet the functional requirement. The two most obvious are the ability to supply the necessary voltage and amperage. But it doesn’t stop there. A CR123A battery could meet these requirements but wouldn’t be able to provide power to our remote. Why? Because there’s no way to connect the poles to the battery compartment of the remote. So, we have another critical characteristic of the ability to connect to the terminals in the remote control. We could have some smarty pants sell us that CR123A battery with leads that clip to the remote terminals, so we’ll go ahead and add that the item needs to fit into the battery compartment. And while we’re at it, a minimum electrical capacity since we don’t want to go through this every day.
Here’s where CGD gets interesting. Did you notice how many critical characteristics were generated? We just need a couple of batteries! And if they don’t provide enough power, we can just change them. In this case, the required quality of the procured items is extremely low. Our acceptance criteria need to reflect this reality, or I’m going to start testing and measuring every battery I get out of a box before putting it in a remote. Luckily, there is already a battery industry standard for size, voltage, and minimum capacity. As long as I verify that I have some AA batteries in my hands, I can accept this item as meeting my critical characteristics. As an added bonus, I’ll add a special test to check that my remote works after installation by increasing volume.
In this case procurement closeout just involves sitting down and relaxing. In more detailed procurements, we might document the checks we did, retain paperwork, take pictures to verify installation, etc. Feel free to do that for your remote batteries, but I can’t be bothered.
This is probably the least well performed step in any CGD process. The situation I see most frequently is an engineer that regularly operates in a highly regulated industry designs some system with an expectation of a universal high quality level in all components. That’s a fine concept, but in many cases this results in severely limiting the availability of suppliers, increases costs, and extends schedule. This is about the point that some downstream manager realizes that commercial dedication can provide many benefits in expanding the potential procurement to lower priced and more readily available items.
The manager charges the Quality Assurance (QA) department to perform CGD on a commercial item to fit into the system the engineer designed. As an engineer, I’m more than happy to let the QA department deal with the headache of paperwork – as long as they determine my design intent first. Too often the QA department fails in one of two ways; either they determine that the whole item has to be as good as an item produced under the applicable quality program, or they determine the critical characteristic based on their understanding of common usage for that item.
The first method of failure results in an extremely heavy CGD process that is set up for failure. The result is either a complete inability to meet requirements by the supplier, giving up on trying to verify requirements by the buyer, or the need for the buyer to accept a great amount of risk due to uncertainty in the item’s ability to meet all documented requirements. What makes it worst is that compounding a standard procurement with copious quality requirements results in much higher costs. This can erase or reverse any benefit to be gained by CGD.
The other method of failure usually results from complacency for a procurement that closely resembles one from the past. Often the QA department will assume that when the same component comes across their desk, they can recycle an earlier dedication process. This thinking does not take into consideration that most components can be used in multiple ways, and perhaps in a unique way.
In both cases the most effective means of understanding the actual design function is to – you guessed it – ask an engineer. Preferably the original designer to understand the design intent. That puts the item in full context of its purpose and allows us to capture what it really does. I have a great example I ran across that I’ll modify somewhat to protect the innocent.
I was reviewing a procurement for ¼ inch bolts made of stainless steel. Some of these bolts held together a structural frame and some of them held minor brackets for convenience purposes in a corrosive environment where a buildup of corrosion products is undesired. The function of the first set of bolts is to be strong enough to ensure the structure is stable. The function of the second set of bolts is to not corrode. Other than geometric requirements, which they both have so they fit, that’s about it. To “simplify” the procurement the QA department required that both sets of bolts can be fully qualified to meet the applicable standard for ¼ inch stainless steel bolt as required by a highly regulated industry. The bolts were to be procured from a supplier that does not have a quality program as rigorous as this highly regulated industry. I think you see where this is going.
Critical characteristics are the attributes of the item that show it meets the functions. Generally, this includes geometry, materials, color, etc. All critical characteristics should be associated with some functional requirement for the item. At this stage, often people will come up with critical characteristics out of thin air. Or so it seems. What is really happening is that there is a function that they didn’t document in the previous step. This is the perfect time to update that function list. One example I see all the time is exclusion of the color yellow. In this case, the function is that the part can make it through a controlled area without getting impounded. I’ve never seen this functional requirement documented but see it as an acceptance criterion all the time, so it has skipped several steps in the process.
Let’s get back to our bolts. Setting aside the dimensional requirements, which they both have, let us consider the unique functions. If we were tying critical characteristics to the structural bolts, I’d have to say that “it’s strong” is certainly one of them. Specifically, how strong it needs to be and in what way. The bolts in a corrosive environment have a critical characteristic like “doesn’t corrode in that environment” where the environment is defined. In our situation with an overzealous QA department, the critical characteristics are the same for all bolts as though they could meet the requirements of a bolt supplied by a regulated supplier held to an industry standard. This prescribes markings, materials, strength, and the list goes on. I think it’s obvious that the critical characteristics are being overly specified, but this ensures the risk is low to the QA department since they are essentially enforcing all the regulated supplier requirements.
The acceptance criteria flow from the critical characteristics. There is great debate about what is an appropriate means of acceptance in different situations, and most regulated industries have their own rules. I won’t delve into those, specifically because I’m only expert in one industry in this regard. Just like all critical characteristics are derived from item function, acceptance criteria are derived from critical characteristics. The result is some set of inspections, tests, analyses, surveys, etc. that provide documentary evidence that the critical characteristics are captured by the procured item.
The importance and difficulty of this step cannot be overstated. The importance of the function for specific items must be taken into account at this point, so getting involvement with the design engineer is necessary. This will drive a lot of the acceptance process, including the acceptance methods chosen, the sampling rate, qualifications of those doing acceptance, rejection requirements, etc.
In our case with the risk adverse QA department, the list of acceptance criteria would be immense. Each of the critical characteristics must be verified, so that requires a documented inspection of markings, material testing and traceability reports, strength testing of sample lots, etc. If we were to reflect back on the original functions, our expectation to this process may be that we have the following requirements:
Note that we’re not assuming “bad actors” and that they mixed in plastic bolts with the hope we’d only select the metal ones for testing. If we were concerned, we could include a survey of their facility and the packing of the bolts to show that they’re all coming off the same production line.
It’s not worth going into too much detail here. The details are determined by the quality program, company procedures, type of contracts, and many other factors. But as in all quality environments, the important aspect is to capture what was done, why it was done, and how it could be done again.
Commercial grade dedication is no different than what we do every day, both professionally and in our personal lives. We determine what is needed based on function, how that function is met, and some objective evidence that it can meet the function. The amount of rigor we put into the process is informed by the importance of the function. Even though the basic concept is straightforward, it is often lost to engineers mired in the requirements, procedures, processes, and documentation generated by the quality standards we work with every day.
As an engineer, it is our job to see the forest for the trees. Application of formal quality programs and procedures can seem overly burdensome and unnecessary. It is important to realize the benefit that can be gained from a process so it doesn’t just feel like we’re turning the crank. By looking at the “why” we are doing these types of activities, we can be more effective, efficient, and potentially provide opportunities to improve the understanding of those around us. Commercial grade dedication should not seem like some foreign activity that is the domain of the QA department; engineers have a vital role in shaping the process and providing opportunities for tailoring that are beyond the scope of people external to the design process.
I’ve used up my quality aphorism already, so I’m going to come up with a new one. So many things come to mind, but if there was one concept that I want to put into a punchy little quip, it’s that commercial grade dedication is no different than the process used to procure pipe from an NQA-1 supplier; you still determine function (to withstand a certain pressure), define the critical characteristics (manufactured under the control of an NQA-1 quality program to a specific standard), determine acceptance (supplier has an approved NQA-1 program and provides a certificate of conformance to the standard), and finalize the procurement (document the part number, where it was installed, how, etc.). As engineers we need to understand this process so we can best leverage it to ensure that our work is properly characterized in the final product. The increasing focus on how CGD is different than the standard quality process is what trips everyone up and creates this sense of dread. Stop being afraid of CGD! You’re doing it already, you just don’t know it.
The quality program serves the needs of the engineer, not the other way around.