Mettler-Toledo International Inc. (MTD) Earnings Call Transcript & Summary

Hi, everyone. My name is McKenna Schultz. I'm the marketing program specialist here at Mettler-Toledo. Thank you so much for joining our webinar today on what is weighing terminology and proper weighing terminology and also just the importance of weighing accuracy and GWP. We're going to be kicking things off in just a minute here, but I want to run over a couple of housekeeping rules real quick. If you guys could make sure that your mics are muted and also your cameras are off just to block in the background noise and also bandwidth purposes, that would be great. As we're going through our presentation today. If you have any questions come up, feel free to pop them in the chat and either someone will help you out in there or we can cover it in our Q&A session at the end. I'm excited to introduce Doug Dodridge. He's one of our product managers here at Mettler-Toledo; as well as Dave Cerullo, who is one of our national account managers. And with that, Doug and Dave, I'll let you guys go ahead and take it away.

Good deal. Well, thanks, McKenna. And hey, good morning, everybody. Thanks for being with us. Welcome to the what is weighing accuracy and proper weighing terminology webinar put on by Mettler-Toledo. Today, we're going to be digging into properly defining some weighing terminology and talking about how this helps us assure measuring accuracy over the life cycle of our weighing equipment. And as we progress today, we really want to talk about what it means to partner with Mettler-Toledo in your weighing quality program and really less about the performance and accuracy of our equipment. Our hope is to kind of make you better or help you better understand how you're weighing equipment works so that you can make some sound business decisions from the data that you're getting from your scales on a day-to-day basis. And this looks like a lot of different things. It might look like more efficient ways. We're looking for more efficient ways to handle inventory control or raw materials. We might be talking about applications like net content control or managing waste or improving yield. Whatever the application or the reason for using a weighing device, it's important for everybody to remember, scales are a critical control point of your business. Proper knowledge of what accuracy really means to you and how you need to enforce it to ensure quality, safety and regulatory compliance is really of the utmost importance. Okay, jumping into the agenda here today. First things first, we're going to get aligned on a little bit of weighing terminology, take a look at kind of what accuracy is, what it isn't and some other terms that play into that. From there, like McKenna mentioned, Dave is going to take the reins and talk more about assuring measurement accuracy over the life cycle of our weighing equipment before we wrap up with some Q&A. And before we move on to that Q&A piece. McKenna did mention the chat is active today, and Dave and I will be supporting each other in the chat as we move on through the session. So if you have questions, during the first portion of the webinar today, feel free to go ahead and add those to the chat, and Dave will do his best to answer questions as he can. Then as Dave's presenting, I'll take over in the chat, and then we should have some time to come off of mute and answer any questions or discussed at the end of the session or pick up anything that we missed in the chat along the way. All right. So with that being said, we'll go ahead and jump right into the content here. First weighing term that we're going to look at is resolution. So resolution is the number of millivolt increments, a terminal or indicator slices the incoming voltage signal into. In this example, we're looking at a voltage span of 5 volts, and we've got 10,000 points of resolution. So that means that every division size or have -- or every half a millivolt that we work our way up to span represents 1,000 points of resolution. When we're talking about resolution, in other words, we can think about this as the number of divisions applied to an incoming weight value, okay? So a scale or a load cell or a weighing device is sending a mill able signal back to our terminal and the terminal is dividing that signal and the indicator is dividing that signal based on the resolution that's been set. Moving on to division size. So division size is the size of that millivolt increment a terminal or indicator slices the incoming voltage signal into. So in this example here, we're still looking at that same 5-volt span that we were looking at on the last slide, but each of those division size is half millivolt. And again, that represents a 1,000 points of resolution. It's important to remember when we change resolution on our indicator, we are effectively changing the division size as well. So keep that in mind. Moving on to readability. Readability, this is a term that often gets confused with accuracy. Folks typically look at a readability spec for a device and use that as an accuracy specification, but in reality, what readability really is, is the weight value of the millivolt increment, a terminal or indicator slices the incoming voltage into. This is always going to be a multiple of 0, 1, 2 or 5 and it's going to round down or it's going round to the larger increment. Oftentimes, this is going to be abbreviated as D4 divisions. You'll see this represented as D in a lot of data sheets and manufacturer specifications and things like that. In this example, we're looking at a 30-kilogram device that has 10,000 divisions of resolution applied to it, which would typically give us a readability of 3 grams. But again, we mentioned readability is always going to be a multiple of 0, 1, 2 or 5. And so in this case, that readability is going to round up to 5 grams. So our 30-kilogram scale at 10,000 divisions is going to have a readability of 5 grams. In general, readability is exactly what it sounds like. It's the visual acuity of a measurement device. And for measurement devices, this is described as the smallest numerical increment that the device can reliably read. I'm going to emphasize that reliably piece that will come into play here in the next couple of slides. Our users and our customers, they often look to increase precision or readability of a weighing device in hopes for a scale, in hopes that it's going to increase the overall accuracy. It's important to remember though that increasing the readability of a weighing device can often have adverse impacts on accuracy if repeatability specification or re-repeatability specification of that readability is unknown. So understanding these 2 specifications in simple terms and knowing how to interpret them is really critical to weighing quality management and more on repeatability next. Jumping into the repeatability piece of things here. Repeatability is the ability of the weighing device to display the same value for the same object or the same mass under the same conditions over and over in repeated setting. This is reported as a standard deviation of the repeated weighing. So in this example, we've got a 30-kilogram scale. We've used the 15-kilogram test weight and you'll notice that a couple of those tests showed some deviation, which is why we've got the standard deviation figure that we see on the screen. Again, the output of this readability test or excuse me, repeatability test as the standard deviation is one of the factors that we use to calculate measurement uncertainty during calibration. We're going to talk a lot more about calibration, measurement uncertainty and how these factors apply to both of those as the session progresses today. When it comes to a repeatability test, we typically see this taken as 10 [weighments]. So typically, like you're seeing on the screen here, we'll typically see that as 10 measurements. However, that can sometimes be more, and that can sometimes be less depending on the device or balance that is being calibrated and what guidelines it's being calibrated to or tested against, okay? So time repeatability and readability together, repeatability is a metric that we use to determine how well a measurement device performs at a set readability or increment size. It's critical to know this metric for a weighing device, both before you buy it, but really after you install typically smaller readability and repeatability values are often associated with higher performance devices and higher cost devices as well. So we mentioned on the last side -- the last slide, the repeatability often changes depending on that readability setting. So it's important to remember that when we're thinking about repeatability, the result of that test is specific to how the scale has been set up when that test was conducted as far as capacity, division size and readability. If we change those things during a calibration in the future, something along those lines, we should expect our readability specification to change as a result of that. So keep in mind how repeatability and readability play into and off of one another. When we talk about precision, precision is the readability of a weighing device, accompanied or when it is accompanied by a repeatability -- or excuse me, yes, repeatability. A great way to visualize this is by using an eye chart that we see here on the screen. The line where we can no longer read all the letters is the precision limit of our eyes. So for me, that's the line on the letter side of things, starting with F and on the numerical side of things, starting at about 0.05 kilograms. That's my personal precision limit of my eyes on the screen that I'm seeing here. The increment size where the scale starts to make repeatability errors would be the precision limit of the scale or of the weighing device. We mentioned reliably a couple of slides ago, and we emphasize that. When it comes to precision, the word reliably really does make all the difference now when it comes to separating the capability of 1 scale or 1 measuring device to the next. Increment size is simply just the setting in the terminal or indicator, and that's something that can be easily manipulated by our service groups, by our maintenance teams, process teams, things along those lines. Anybody that has access to those terminal settings may have the ability to manipulate those things. Just like we can easily change the font size, like you're seeing on the screen here from large to small, we can do the same thing with our increments in the indicator. Being able to display that value is a lot different than being able to read it and oftentimes the same can be said for measuring devices. So again, when we're talking about precision, it's readability when it is accompanied by repeatability and the precision limit of that device is really met when we start to make repeatability errors at the readability that's been set. I mentioned a few slides ago that we use the output or the standard deviation from a repeatability test as one of the inputs for the calculation of measurement uncertainty. So now we're going to talk here about measurement uncertainty. And really, the idea here is that it's the amount of error in a measured value on the indicator or terminal. It's important to remember, regardless of the device that we're using, what we read on the display is simply an indicated value. Every single measurement that we take on every single measurement device, no matter what type of load cell technology we're using, no matter how precise that device is going to be is going to come with some level of uncertainty. At the end of the day, it's our job to understand what you're looking to achieve, what you're looking to enforce from a quality, safety and regulatory perspective to make sure that you stay compliant and then ensure that the device that's being quoted or sold is fit for purpose within those guidelines and within those requirements. In this example here, what we're saying is the indicated value is 100 kilograms. But based on our understanding of the weighing device, the true measurement could be anywhere between or the true weight value could be anywhere between 99.5 and 100.5 kilograms. Now for some customers or for some users, they may say, "Hey, that's fine. That amount of uncertainty or that tolerance band doesn't compromise our process at all, again, from a quality, safety, regulatory side of things and keeps us compliant. However, some other customers or some other users, other applications must call for much -- or do call for a much tighter tolerance because maybe this level of uncertainty is unacceptable. So this is what we look to really have good discussion around and conversation around as far as what your expectations are and what type of accuracy you need to enforce, what type of uncertainty is acceptable to ensure that the device that's ultimately -- the device that's ultimately used and installed is appropriate and fit for purpose. And when we visualize measurement uncertainty, there's a couple of different ways that we can look at it graphically. First is in an absolute sense, and that's expressed as a unit. So here, we see an absolute measurement uncertainty curve. And as we move up the span of the scale or as we move up towards capacity, you'll notice that the absolute uncertainty expressed as a unit increases, which makes sense. We would expect that the uncertainty applied to a higher value would be higher in correspondence to that value than it would for a lower value in the lower end of our weighing range. We can also express measurement uncertainty in a relative sense as a percentage. So here, we're looking at a relative measurement uncertainty curve. And you'll notice at the low end of our relative measurement uncertainty curve that uncertainty is much, much, much higher. So when we talk about measurement uncertainty, and we talk about tolerance, we want to make sure that we've got a good understanding of your tolerance, what's acceptable and what's not so that we can use that to plot against the relative measurement uncertainty curve and to determine what the accuracy limit of that weighing device is going to be. We refer to this as the minimum weight of the device. So we talked a few slides ago when we were looking at precision and we were looking at the eye chart about the accuracy limit of our eyes or the precision limit of our eyes. Here, we're looking to do the same thing through again, understanding the tolerance that you need to enforce either from a regulatory or from a process standpoint and then apply that or plot that against our relative measurement uncertainty curve to determine what the operating limits of the scale are going to be. The idea here is that if we're weighing above the minimum weight, we've got good confidence that we're able to enforce whatever the stated tolerance is across any measurement up to the capacity to scale. But if we weigh below the minimum weight, there's a high likelihood that the uncertainty applied to that measurement is going to be beyond our acceptable tolerance and could cause problems for us later down the road. So really go ahead and button up all of these terms by defining here what accuracy really is and what it is not and how that pertains to you. So at the end of the day, close enough equates to accuracy. It depends on the users or the customers' needs, and it really has nothing to do with the level of detail or digits beyond a decimal that we're reading on a display. We've established that no measurement is exact, meaning that there is some level of uncertainty in the number that we see on the display. In this example, again, we see that 100 kilogram, our measurement uncertainty is 0.5% based on what we've understood from calibration. And so our value could be anywhere between 99.5 up to 100.49 kilograms. The user of the weighing device has to decide how much uncertainty in that number is acceptable before it becomes an unuseful measurement. This is what we call process tolerance or weighing tolerance. So when we refer to the tolerance that we're looking to understand for your business or your process, that's what we refer to here as process tolerance. Accuracy is a condition in which a measurement has less uncertainty than the process tolerance permits. So in this case, the amount of uncertainty that will be applied to the measurement that we're taking of 0.5% is less than the tolerance that we're allowing for application, deeming this an accurate measurement. On the inverse, if the tolerance that we were looking to enforce was tighter and was that 0.5%, but the measurement uncertainty that we came to expect from the scale was up to 1%, then this would be deemed an inaccurate measurement, because the uncertainty exceeds the allowable tolerance that we have for our process for our application. All right. So that gets us through the weighing terms portion of things. I think Dave is going to take over control here and do the next portion of the session, but we'll just take a quick check and see if we've got any questions in the chat.

I do not see any questions in the chat at this time.

It doesn't look like it. So thank you all for your attention during the first portion. Dave, I'll flip it over to you, and then we'll have some additional time for questions at the end.

Sure, Doug. And go ahead and stay on the camera here for a second and just let me know that you have -- you can see my presentation, hear me loud and clear. Good.

I see your presentation, I hear you loud and clear. You're good to go.

Great. Thank you, Doug. So folks, thanks again for attending today. Doug just did a very nice job of telling us, informing us about the science of weighing accurately, the science behind metrology, how -- why it's developed to assure weighing quality over the life cycle of the scale. The next portion here is considered more of the applied science. How do you, as the end user? How do you, as your -- as the customer or the manufacturer of a product or the analytical testing laboratory? How do you apply what Doug has just told us in your everyday application? So with that said, I just want to make sure that we are inviting anybody on the call no matter what your industry, whether you're in food and beverage, fine chemicals, chemicals, pharmaceutical, whether you're in laboratory, if you're in pharmaceutical laboratory, you have questions about the USP or general Chapter 41, General Chapter 1251, if you're ISO 22000, if you are concerned about the global food safety initiative, ISO 17025 accreditation, any industry. We're ready to help you. Please feel free to use the chat and ask us questions about this applied metrology and how it may affect your application, okay? So let's get into it. Using some of the topics that Doug has just described. What we want to do is fundamentally control our measurements or as Doug said, assure measurement quality over the life cycle of the scale. And if we look at any of our regulatory documents or quality guidelines, one of those first steps is the plan or design qualify. We look at exactly what we are intending to do with the measurement device, in this case, a scale or a balance and then select the most appropriate scale balance that is fit for purpose. And we don't do this in a methodology like we used to in the past. Even at Mettler-Toledo 10, 15, 20 years ago, prior to our weighing quality program called GWP. A lot of times, our purchases were done by history and habit. So often, our personnel on the phones or our field sales representatives or field service representatives would receive calls me included as a service technician for many years, get a call from a customer, we want to buy a new scale, 300 kg to 0.001 kg, referring to the division or increments that Doug was referring to before or maybe we want to buy a new balance. We've been weighing -- we want to weigh some milligram quantities. We need a 200-gram balance. We've got a volumetric flask we need to use 200 grams to 0.001 gram or 0.1 mix. Okay, stainless steel, preferable on the scale, possible USP monograph substances for the balance. Sure. Absolutely, our sales reps would say, send me a PO number, I'll shift these right out to you. They'll be there Monday, maybe not in the supply chain world that we live in today. But call me if you want to have a tech come out and help you install it and by the way, do you have any weight you could throw on it once in a while to make sure it's accurate, okay? History and habit purchase, we're leaving a lot of open or open areas for conversation or gaps in this design qualification portion or this planning portion. ISO 9001 uses the graph PDCA plan, do check and act. And of course, any good pharmaceutical company that has a weighing quality program will have an equipment qualification mechanism, starting, of course, with the DQ, the design qualification. So a history and habit purchased by just simply purchasing on capacity and division size is not considered a best practice. So what do we do? We take a look instead of going to just capacity and division size, we analyze what the customer, what our customers want to measure. And in this case, we're looking at a scale, an industrial scale for batching. And if you notice the total number of compounds is 6 and the total gross weight of this batch, the actual batch, not the tear container, the batch, the compounds totaling 100 kilograms. And our customers, what we ask to our customers is what are you looking to achieve even though as it's stated here, you've purchased a 300 kg scale, the 0.001 kg last time to measure all these compounds and you do throw a few weights on it once in a while, it occasionally, or maybe never fails and we have a service provider coming in once a year for maintenance calibration. So I think we should be good just to buy the same scale. But what we do with Mettler-Toledo today is we push the time out button or the pause button, and we want to understand more. And I want to call special notice to you this little note down here at the bottom of the screen, ladies and gentlemen. I've written many weighing SOPs for many different customers in many different industries. And one of the topics that is most often disputed is that if I put a drum on a scale or if I put a volumetric flask on a balance and I've exceeded that minimum weight that Doug spoke of earlier, that accuracy limit. I've got the drum on the scale. Now I can measure anything. Well, best practices and science even in some of the calibration guidelines and even in the USP state that every net measurement, no matter be a formulation or a filling operation, every net measurement needs to be larger than the minimum weight of the scale. And the net weight does not include the gross or the tare. So you can't exceed or be past the minimum weight of scale with a tare vessel. So please keep that in mind. So in this case, back to this example, the 2-kilogram measurement is the smallest net measurement. So when you interact with Mettler-Toledo, whether it be on the phone or in person, whether it be direct or one of our distribution partners. They will ask you a very important question, what is the smallest net measurement you are looking to weigh on this scale. Okay? Please keep that in mind. Again, taking Doug's information and applying it okay? So that is the smallest net weight of this batch with this formulation, okay, or this recipe. Then, as Doug alluded to earlier, we want to apply the weighing tolerance or the process tolerance, how much error are you willing to accept even if as the example Doug showed, even if that measurement on the scale is indicated as perfect, if it reads 2.000. Doug clearly told us that no measurement is perfect, and it could be because of rounding, it could be because of systematic deviations, the repeatability test. Whatever that is, we need to allot for some type of measurement uncertainty and is that in a relative form, as Doug was talking about, is that in a percentage value or is it in a quantified value plus or minus 0.020, is it 1%? What is that value that your process could accept without compromising quality, safety or an extravagant business loss. Think about that. And again, it ties back to the previous slide. Writing SOPs for our customers, it's the second most questioned or mostly discussed topic, when we write weighing SOPs, it's around the weighing tolerance, also known as the process tolerance, okay? So we ask those questions at the design phase. What do you intend to use this device for? We then don't hand our customers a data sheet. Well, we've got the linearity expectations. We have the repeatability expectations. We have the error of indication expectations in typical settings. That's a data sheet. What we do is we take the information that you provide the process information, that planning or design information and we link it to empirical data that we have here at Mettler-Toledo to design, qualify. This document that you see on the screen is not a data sheet. It is a design qualification tool that our sales reps, both direct and distribution partners have access to, to design qualified balances and scales for our customers no matter what the industry, okay? So Doug, I'll pause there to see if there's any questions in the chat about selecting or planning or designing a new scale purchase. Okay?

So not seeing any yet, Dave.

Thank you, Doug. What we want to do next, of course, is we buy it, the scale hits the loading dock, the balance gets into the laboratory. The first thing you want to do, of course, is unpack this thing. It's brand knew, I just spent some good amount of money on this. I want to start using it. The importance of installing and calibrating a new scale after purchase is another step. So it's the installation qualification, of course, we have the check boxes in our installation packages where we say, yes, all the peripherals work. Yes, it's set up in this configuration. Everything is set. We have all of these things that are shipped, okay? It's installed. It's communicating. It's interacting with a PLC, it's printing, et cetera. That's the installation qualification, of course, and then we go to the operational qualification. That's the calibration. We've installed it. Now we want to see how it performs. We've design qualified it. It should work. Now that's proven with the installation and operational qualification or in ISO 9001 is the plan. And then we do the PD, the do the check, the TDC plan do check, that's what we're doing here when we install a new scale. It's the installation and this is me Dave Cerullo on the bottom here, talking to my customers, circa 1997/2002 as a service technician, installing balances and scales. I've installed your scale, calibrated it. Here's the before and after data. It passes manufacturer specs if it's a balance. Here are the calibration reports. Yes, you might want to throw a few weights on it once in a while. Thank you for your business, see you in 6 months. But what we did over the last 15, 20 years here at Mettler-Toledo is we've changed our interaction at the service level and the installation level with our customers to ask questions. So we don't leave our customers with the questions in the blue field. Well, we've always tested our scales in a certain way, why change now? Does measurement uncertainty come with the installation? Do you calculate that? Or do we? How about ISO? What -- you mentioned weights, what weights do I use? USP does General Chapter 41 or 1251 apply to my manufacturing scales. I see the data output from your calibration is different from my internal calibration department. Why is that? Why is it different? So we don't want to leave our customers with guess work. So the first thing we do at the point of installation, again, Mettler-Toledo Direct Service or our distribution partners is we install the scale properly. The engineering of the scale is of paramount importance during installation. Is it shined correctly? Is it level? Is there a gap between the deck and the roll on ramp. This information is of paramount importance to us in the installation process. A balance, is it level, is it sitting next to a centrifuge. Is there a huge air duct over top of the balance that is calling -- causing environmental disturbance and affecting the measurement quality. All of these items need to be considered during installation. And then once we have it installed and it's set to go, we feel good about the engineering and how then the environment that it's installed in, we've mitigated environmental impact on the measurement quality. We calibrate. So what you see here on the screen folks is an example of a calibration certificate. Doug mentioned it earlier, measurement uncertainty, a true calibration within the guidelines of Euromet or the ASTM contains a statement or an estimation of measurement uncertainty, and this is critical. Again, no measurement off is perfect. So if we understand how far off a measurement could be. We can then draw a relative measurement uncertainty curve and establish a safe weighing range tying directly in applying what Doug has stated earlier about using measurement uncertainty to establish relative uncertainty. And if we know your process tolerance, the y-axis here, ladies and gentlemen, if your process tolerance is 1%, we can establish the minimum weight on the scale as 1 kg for this example, and this is the applied science. And of course, this safe weighing range, if I establish today that the minimum weight of the scale is 1 kg, what happens if my service department comes in, in 6 months, and that measurement is -- that minimum weight is 1.3 kg. So what we do is we establish a safety buffer. We maybe multiply the safety factor maybe a 2 or 3 week to establish a safety buffer. So if that minimum weight or the accuracy limit of the scale fluctuates over time, you're weighing safely up here. So we would say measure net measurements at higher values than the minimum weight of the scale to assure measurement accuracy over time, okay? So what happens after installation? That's great, Dave. You've engineered it. We've got it all shined up. It's level. The balance is not getting affected by any turbulence from wind or heat or vibration awesome. What happens next? Well, that's where we really go into the applied science. We have the operational qualification, of course, ongoing calibration and the performance qualification and even maintenance qualification. What happens next? So we look at this a little bit of metrological accuracy and just tying back in what Doug was talking about earlier, is the device true and repeatable? Is it true? And is it precise? Can we measure a value, whether it be a mass or whether it be an unknown, a powder or liquid, a component, can we measure that? Can we get a true value? And can we repeat that measurement? If so, then this device is true and precise. So that's metrological accuracy in a nutshell. What we then do is we apply. So taking what Doug talked about earlier and applying it to process accuracy, as stated understanding if the measurement uncertainty of the device is less than the weighing process tolerance or the weighing process requirement, again, applying science to your everyday weighing where measurement application, whether it be a balance or a floor scan, okay? And again, how do we do that? ongoing operational qualification or checking in ISO 9001. We checked the device. We calibrate. So Mettler-Toledo while there's no user manual that prescribes how often to calibrate or maintenance a scale should be based on risk, Mettler-Toledo does suggest a once-a-year calibration and maintenance event by an third-party external accredited service provider that can execute calibration on scale. So again, it's a best practice to have it at least once a year to understand the performance of the scale. And as Doug said, the output is a calibration certificate and a statement of measurement uncertainty. And that looks like this. In the Mettler-Toledo world, we have a certificate called the ACC, and we can estimate measurement uncertainty throughout the entire range. And folks. I know it looks -- it's a little bit of science and may look a little complicated, but what we can do here for our customers in any industry, the measured value of anything you want to measure, we can plug into R -- this value R can be removed and we could plug in 5.64 kg. We can plug in 3.19 grams. What the measured value is we can put it in place of R, and we can estimate or quantify the measurement uncertainty of that measured value, okay? And then what's really cool is that we can give it in a relative form. So yes, 3.15 grams maybe plus or minus 113 milligrams, but relatively speaking, that's 3.2%. Interesting. So again, we're applying this science to assure measurement accuracy over the life cycle of scale. And I get this question all the time. So what does that mean for me, Dave? That's great. I've checked the box, calibration, measurement uncertainty, I've checked the box, what do we do next? How do we use this? Well, what we do, again, is take this formula and we can quantify it. The material weighed is x, the absolute uncertainty is the yellow portion of what you see on the scale here. So yes, 35 kg on the far right-hand side here, we look at it, we'll measure 35 kg. The indicated value is 35.00 kg, but no measurement is perfect. -- that might be off by plus or minus 1.83 grams, okay? Dave, is that a lot? How about my 15-gram measurement? It's off -- it could be off by 0.114 grams. Is that a lot? Well, hang on a second. We're just quantifying it here. We're telling, we're giving you an estimation of uncertainty, 15 grams, plus or minus, 0.114 grams. Well, is that a lot? Well, then what we do is we take the absolute uncertainty and divide it by the mass and we get that relative equation. So is -- that's relative, okay? So is that a lot, Dave? It's relative. Let's take a look at it another way, okay? The relative measurement uncertainty, just a quick refresher. As you get into the lower end of the measurement range, the relative measurement uncertainty explodes. It gets so high that you may not even be able to use the device to measure accurate -- accurately at the low end of the measurement range. So again, relative uncertainty established from calibration, okay? So again, is that measurement being off by 0.114 grams a lot? Well, it could be, but it's a comparison about what you're willing to accept. Is your process tolerance 1%? Well, that? Guess what? The relative uncertainty is smaller or the measurement uncertainty is smaller than what you budgeted for. So yes, that measurement is okay. Again, even if it indicates 15 grams perfectly, it could be off by 0.114. You established that, that's okay because it's less than 1%. But at the low end of the measurement range down here, this 20-milligram measurement. And while this does not seem like a lot, it's off by 0.113 grams, relatively speaking, it's off by 565%. Nobody is going to establish an uncertainty budget of anywhere close to that. Most often, ladies and gentlemen, USP 41 dictates 0.10%, most often, in most other industries, a manufacturing or an in-use tolerance is somewhere between 0.5% and 2% or maybe even 3%. So please keep that in mind. USP 41 dictates 0.10. Most customers in pharma, manufacturing and other industries use an uncertainty budget or process tolerance of 1%, okay? So again, we're applying what Doug has put in force before. Relative uncertainty, controlling the measurement don't just check the box that calibration is done, work with Mettler-Toledo or our distribution partners in lab and in industrial to understand your process requirements. This is a big deal. Why? Because very few service providers can deliver this information. We use calibration to control your weighing process. And another important piece of this, if we look at ISO 9001 when we look at USP General Chapter 1251, they state or require evidence of fitness for purpose of measurement devices. So if we define the criticality of your measurement devices, we understand what they're being used for, the process requirements as you see here, the user requirements, okay? Then we compare that to the achievements from calibration. We understand if this device is fit for purpose, the installed device is fit for purpose. And ladies and gentlemen, we can execute this type of service for any scale or balance, whether it be a Mettler-Toledo or non-Mettler-Toledo scale or balance, we can evaluate if that device is fit for purpose as compared to your weighing requirements. Great day, that's awesome fitness for purpose. I got this thing. I'm hung up on what ISO 9001 is saying, it says, calibration, performance verification or both at specified intervals and it's speaked to be established by me, the customer. There's no guidance on what weights to use, what tolerance to use? What is that interval how often should I be testing my scales? How frequently should I have calibration? Well, if we take a risk-based approach like Mettler-Toledo does, we established calibration and performance testing through the impact on an inaccurate measurement. If that impact is going to impact quality -- excuse me, that measurement is going to impact quality, harm humans, harm animals, harm the environment, if it's going to impact your business dramatically, monetarily, the Y-axis, if you have a really tight process tolerance, if your quality guideline, can't accept more than 0.1% or 0.01%. We need to have calibration and performance testing more often. A risk-based approach, much like the ISO 9001 and 2015 guideline dictates. Dave, what about testing? There's a lot of information here on the slide, ladies and gentlemen, but we tell us -- we tell our customers don't test to your process limit because if you fail, you stop, your failure equals an investigation. You have to open up an investigation because product might have gone across that scale and got into the consumers' hands without you knowing it. So what we recommend is testing, taking your process limit and squashing that maybe a 4:1 ratio and doing testing, performance testing at this value, even if your process limits are out here. We never recommend down here manufacturer specifications or using Table 6 of Handbook 44 outside of legal for trade applications. So not applicable. And Handbook 44, Table 6 not applicable for pharmaceutical manufacturing, applicable if you're selling goods for monetary exchanges on a scale, absolutely. Again, Mettler-Toledo recommends establishing the process limit and then doing performance tests within a certain value of 4:1 ratio. And last but not least, ladies and gentlemen, the question that we just went through, how often should I test, Dave? What weights should I use? Should I test eccentricity repeatability, sensitivity? How often should I do that? Which weights should I use? You mentioned throwing weights on it? What class of weights? How often should I have my weights recalibrated? Mettler-Toledo can provide clear guidance not only to establish if your installed scales are fit for purpose, but also how to monitor them, check them, performance qualification over time to assure measurement accuracy over the life cycle of the scale. And with that, ladies and gentlemen, I will say that this is something that Mettler-Toledo developed in 2007. It is the good weighing practice program for life cycle management of weighing equipment. Doug Dodridge, myself, we are -- we drive this program within the industrial organization and our laboratory partners. We've trained our distribution partners and our lab sales and lab service and industrial sales and industrial service partners around this quality. We build our people up with GWP. We train our customers with GWP principles. It's not hoke, Mettler-Toledo magic. It is metrology, it's quality programs, and it's your customer process requirements, all baked into a solution to assure measurement accuracy over the life cycle of the scale. And with that, I'd like to stop and open it up for Q&A to see if there's any questions in relation to what we've just discussed today. The science behind it and the apply to your application. And again, ladies and gentlemen, please feel free to come off mute, ask questions live or put them in the chat, any industry, laboratory, industrial, whatever you need to do. Please feel free to shoot away.

We've got a couple questions in the chat, and I'm going to share again, Chuck just asked if we could show Slide 25 again. So I'm just going to pull that up real quick.

So your slide, Doug, or my slide I'm sorry.

It was the calibration, the ACC slide. We had a question regarding accounting application here from Melissa. Talking about using it sounds like currently using an ICS 465 for counting different automotive parts and things of that nature. Hey, typically, we can follow the same type of methodology and we often do for counting applications. Typically, the approach is a little bit different when determining tolerance. We try to understand how many parts we can be off for the smallest part of all the different parts that you're counting, what is the smallest part and what's the weight of one of those parts. And then what's the smallest amount of those parts that we're going to count. And then how many can that be off. In the automotive industry, in a lot of cases, they say it can't be off any or our customers will reject those parts or shipments of those parts. It can't be plus or minus 1. In those cases, there's a couple of different ways that we can address that using some features within the scale. One is called average piece weight optimization and the other is called totalization. And those are both things that are available on the ICS 465 scale that you have currently. You also mentioned looking for assistance in changing the setting from standard reference to a custom amount. That's absolutely a possibility and something that we can follow up with you follow up on later. There's different settings on the ICS scales where you can have a customer -- a fixed reference set of 10 or something like that or you've got a variable reference where you simply plug the numbers, I'd say, 20 and hit the variable button, and that tells the scale that I'm using a count of 20 for my reference. We've got your e-mail Melissa. So if you have any other questions on that, feel free to come off of mute, but I'll follow up with you additional details of the e-mail there.

Doug, I see a question from Chuck Olivier about selection calibration and maintenance of scales for commercial cannabis. Chuck, to answer your question simply, it's pretty much the same application, the applied application, the plan do check act cycle that we just talked about. Using -- you're using any type of process tolerance you can establish selecting the right device, of course. It doesn't -- I'm assuming it doesn't need to be a wash down unless you're working with solutions. But yes, it's basically the same. It's the same process. We don't -- Mettler-Toledo right now -- as far as regulatory because it's not regulated by the FDA. At this point, nationally, we don't have guidance on the regulatory piece, but locally our experts will use the science that we just discussed to help you do exactly what your question is, Chuck. Chuck, do you have any follow-up for that? Doug, would you like to add anything?

I think you got to cover on that side?

Doug, I do see a comment as well that they are looking for the other Slide 25? Are there 2 separate PowerPoints there?

I don't believe so, unless Dave's numbers may have been a little bit different than mine. However, I thought we were working off. Got you. Chuck, I was answering the question when I stated that.

Is this the slide we're looking for?

It looks like it was 1 with some terminology on it.

Yes. Okay. So I think, McKenna, I've shared my screen, I think, and can you see my screen. No. One second I think I got it now. Yes, you got it. Okay. So Chuck's questions on this. Maybe grabbing a picture. Yes. So these are the major tests that attribute to metrological accuracy. There's also another 1 eccentric loading that we do take into -- so we have -- there's 4 components. There's rounding at load, there's rounding at 0. There's eccentric loading and then there is repeatability. All 4 of those parameters are tested through calibration on the scale. So to establish that's metrological accuracy through testing of the system, then to calculate measurement uncertainty, there's a quadratic formula that takes all 4 of those parameters and adds for more parameters from the mass itself. That's a -- that's how we follow and establish measurement uncertainty, and this is not a Mettler-Toledo formula. This comes from the Euromet CG 18 or ASTM-898-20. So this is an international guideline for calibration and it includes these tests, okay? I hope that helps.

I want to make sure we're not forgetting anything. Braxton Wallace, if you're still with us, your question came in right before Dave got to the slides on routine testing and performance verification and things like that. So again, our approach when it comes to recommendation around routine testing and the weights that should be used are a, specific to the device and then b, specific to the risk assessment associated with that GWP process that we walked through as far as what tests should be completed and at what frequency they should be completed. So firstly answered your questions on the routine testing side of things. But if you have any additional clarification, please free to let us know.

Great. And I also -- and Doug, I'd just like to take a real life example. Don Field had a question about her process tolerance, we don't have a process tolerance. Going back to the slide I showed earlier, I just had a situation with the exact question at a customer in Richmond, Virginia, outside of pharma. And what they asked was Dave, I need to establish what my scale is capable of, can you help? Yes, of course, we can. So the first point, Don, is we calibrated the scale. And then what we did, Don, is we took that information and we established a measurement uncertainty. And of course, that we had a -- what we did was we graphicalized that if that's a term we graphicalized it. What is the scale capable of at 1% or at 0.5% at 0.1%. And we put that into several reports, all culminating into this GWP verification document. So again, to answer your question, we reverse engineered it. We didn't know the process tolerance of what they wanted to use it for. They wanted information on what to use that scale for because they've never evaluated their scale. So we could help you with that. We can make several forms of this document, 1%, 5%, 2%, whatever that is, and establish what that smallest net measurement might look like as compared to those process requirements, okay?

That goes hand-in-hand with Melissa's question as well, kind of understanding the accuracy of that ICS 465 that they've got currently and what they can expect out of it. So keep that in mind, too Melissa talked about some different ways that we can address different things with some different features. But through calibration, we can level set on what the current capabilities of the device are today in its installed environment.

Cool. And anybody -- any of our participants, audience using balances or have any questions around the USP general chapters? Obviously, general Chapter 41 being mandatory for the QC lab weighing of analytes, 1251 being a best practice. Anybody out there in the pharma world when I asked any questions about the USP chapters. Okay? Then Doug, I think we've taken care of all the questions in the chat. McKenna, I'd like to say thank you to you for hosting and Doug and of course, my partner and colleague and friend for being here with us, McKenna, audience, thank you for everything on behalf of Mettler-Toledo here in the United States and in Canada. Thank you for participating today. And on behalf of Doug, my friend, of course, as well. McKenna, we'll turn it to you for closing.

Yes. Again, thank you, Dave and Doug, for taking the time for the awesome presentation and just all the Q&A as well. That was a great session. And everyone who attended, thank you so much for your time today. Myself and Christina Johnston have included some links to some upcoming webinars for the U.S. and Canada as well. So if you enjoyed today's session or looking to learn on some other topics, feel free to check out those links. If you have any questions about anything that was covered today, feel free to reach out, and we can get you in touch with the right person or help you out directly. But otherwise, I'll conclude today's session. Thank you so much.

Thanks, folks. See us again.