Visions Podcast: Fixed Scanners, Big Gains

Visions Host Jim Tatum interviews Charlie Long of Zebra Technologies about a fixed-scanner deployment solution at a Royal Canin distribution center that improved safety and increased loading efficiency by 50%.

Related: Total Wine and More Implements Machine Vision, AI to Streamline Store Operations

Visions: A Machine Vision and Automation Solutions Podcast, is the podcast for engineers, designers, integrators, and end users who want to keep an informed eye on the imaging and machine vision industry. Every second and fourth Tuesday we will explore the latest in imaging trends, developments and solutions. Here you will find interesting, useful insights and observations from expert interviews, solo episodes, even the occasional panel discussion, all of which aim to expand your knowledge on imaging and machine vision. 

Be sure to subscribe to Visions: A Machine Vision and Automation Solutions podcast on Podbean, or wherever you find quality podcasts.

Transcript

Well, hello and welcome to Visions, A Machine Vision and Automation Solutions podcast. I'm your host, Jim Tatum, senior editor of Vision Systems design and visions is an endeavor business media production from your friends at Vision Systems Design. Here you'll find the latest on everything from end user machine vision solutions to trends, developments, and perspectives on all things machine vision and imaging. Whether you've been working in the industry for a while or you're just starting to take a closer look at it, this podcast is designed to grow your knowledge and bring greater focus to your understanding of the imaging and machine vision industry. And now on to our show. 
Well, hi everybody, and welcome to Visions. You know, these days it seems the world is just overwhelmed by so many mind-blowing technological innovations and advances geared to solve all our problems. Yet for all the bells and whistles, often the smartest solutions are rooted in simplicity and in the tried and true. 
Today's episode illustrates just that. 
Royal Canin, a French company that manufactures pet products, recently made some fairly simple but effective changes in one of their distribution facilities. In that facility, forklift operators were doing what so many warehouse workers do every day, stopping mid flow, hopping off the lift, scanning barcodes by hand, all the while trying to avoid incoming traffic as they did this. The challenge wasn't manpower, data, or work ethic. Instead, it was finding an answer to one question how do we verify that the right pallet gets onto the right truck at the right time, without putting people in harm's way or slowing everything down? 
Royal Canin turned to Zebra Technologies, who came up with a solution that resulted in a fifty percent increase in loading efficiency with no workforce reduction, no process overhauls, and a return on investment measured in weeks, not years. 
Our guest today is Charlie Long, VP and general manager of Zebra's Machine Vision Business Unit, who will unpack just what the solution is, how it works and why, at least in this case, the tried and true definitely trumps bells and whistles. Charlie, welcome and thanks for spending a few minutes with us. If you would, please give us an overview of Royal Canin's challenge and Zebra's solution. 
CL: Ultimately, if you look at what Royal Canin wanted to do there, they faced a similar obstacle that a lot of manufacturing slash distribution centers face, which is operators having to hop off the forklift to read a barcode. So it's, uh, it's easy to take a handheld and stick it through the front windshield of the forklift and read if the labels are facing the right direction. But if it's on the side or the opposite side, then it's impossible to get to. So, they had operators that were in danger of getting hit by other operators, or they were holding up traffic, or they were taking longer than they wanted to read because they were looking for a barcode on a pallet. So, what they were looking for was a way to automatically read the pallets before they went on a truck. And it was really a pretty simplistic solution. They spent three years looking for a solution for this because ultimately, we didn't get involved in the conversation till later when somebody saw, hey, we've got all these zebra handhelds. I wonder if zebra can help us. And that's where we stepped in and said, yeah, looking at what you want to do, we just need to put a scanner in place that can read before you go on the truck. So that's ultimately what we did to solve the problem was to prevent the operator from having to get off the forklift, or if anybody from putting themselves in harm's way. But that was the real issue they were facing. 
JT: So basically, this is this doesn't really change manpower so much as functionality and direction of things. 
CL: Yeah. Realistically, it allowed them not to have to make any changes in manpower. So, they were able to keep all the same workforce in place, but increased efficiency. Their efficiency increased by just keeping the forklift operators in the seat was a fifty percent load increase. And that's a simple change. That's a huge efficiency gain. The ROI they reported on -- it was like less than three months is what they said. And that I mean, a fifty percent increase is huge. But again, it didn't require them to make any changes. And how many associates they had on the floor. If anything, they could have more associates on the floor because they freed up traffic and are able to move more product so they can increase production rather than having to decrease. 
JT: And this is kind of something that everybody, the reason I'm even bringing it up, is because everybody on the planet seems to be wanting AI for something or other, and it's very useful. But this would be one of those functions that doesn't necessarily need such a thing? 
CL: That's correct. So, in this particular situation, no AI was needed. The sensors that they're using are the scanners that they're using from zebra have AI capability on them so they can do some more advanced inspection if they want to. But they didn't need it in this particular application. 
JT: Well, you just kind of answered that follow up. How, how a fixed standard does change system dynamics. 
CL: Yeah. I mean, it can change in a couple ways too, right? Like one is it gets, it gets people out of the way. But the other one is that a fit scanner has a better field of view and typically a better reading algorithm than a handheld scanner. But handheld scanners are really intended for when you know where the target you want to read is at, and you don't have to hunt for it. Fit scanners, especially when you can run multiple scanners in parallel, allow you to look at multiple places at one time, or look at places that the operator can't reach. That's really where it fits. Industrial scanning makes a bigger difference. It can also scan at a higher rate, cover a larger field of view. You don't have to deal with the same motion blur that you've got to deal with. When you're using a handheld scanner, that is not going to be held perfectly still. So, it really takes out a lot of variables and increases capability of the biggest drivers to go to fixed scanning. 
JT: Well, for very basic breakdown. How exactly is it set up? When you say it's fixed, is it on the forklift or is it on the wall or where is it? 
CL: So, in the Royal County example, it is fixed on a pair of pillars that are on either side of a dock door. So, there are a couple of scanners on each pillar. So, we can get a larger field of view. And then they're scanning both sides as well as the top and in front of the pallet as it comes through the door. You can also mount a scanner directly on the forklift if you want to. That's common application for us as well. That way, if you think about the scanner on the safety gate of the forklift, maybe you've got a read. When the forks are thirty feet off the ground, you can do that very easily with scanner mounted there versus an operator trying to get a vantage point to read from the ground on a pallet that's three racks up. So, or you want to verify what's on the pallet. And this is another thing where we made the difference. At Royal Canin, we can tie the scanner back into one of our tablets. The tablets are typically already tied back into your WMS, so it makes it very easy to get that data through the scanner but get it to the same place you're used to sending your handheld data to. 
JT: And that's all using wireless, I guess Wi-Fi or 5G or whatever. 
CL: That's right. You've got Wi-Fi capability. And in some cases, we're using Bluetooth. But the, uh, as far as the connection from the scanner back to the handheld or to the tablet. And then of course, the tablets can run off of dedicated 5G, commercial, 5G, Wi-Fi, whatever you have for a network inside your facility. 
JT: And it's not analyzing images so much as making counts and identifying what something is. 
CL: Correct. That's right. So, the industrial scanners typically use just to read a barcode, you're going to read a 1D or 2D barcode. Again, I mentioned earlier, we do have AI built into our systems to make them more capable. One of the most common AIs that we use is a deep learning OCR, optical character recognition that allows us to read text, not just barcodes. And we can do that at the same speed and decode rate that we can read barcodes at. So, if you're not fully encoded across your ecosystem, then we can just read text also. But yeah, generally you're not doing a lot of visual inspection or classification. With a device like this, you're typically doing some kind of identification or reading. More likely, you're reading barcodes on pallets that are in a certain part of the warehouse. 
JT: How do they get there? 
CL: Yeah, that's a great question. 
JT: I really don't know. 
CL: No, no, I mean, there's, so there's a couple of different things that you want to look at. One, if you look at the process we're talking about, what we're talking about is when a pallet gets picked from a rack position and loaded onto a truck. And what Royal Canin was trying to verify was the right pallet got on the right truck, with nobody having to do multiple scans to make sure it got there. So, the most important thing they wanted to do was read. As the forklift passed through the door into the truck. They would know this pallet went through this door at this time, and then we can tell what truck was there. What customer that was headed to, the right thing went to the right place. But an even better question is how did the pallet get on the rack in the first place? So, in this situation, they're manufacturing the product in the same facility. So normally the product's going to be manufactured. It's going to get packaged at the consumer package level. It gets scanned at that level as well, typically by its industrial scanner and put into a case or a larger pack. And then those packs are scanned as they're put into a larger case that holds multiple cases that will go on a pallet. So, then you palletize and then that pallet is assigned a pallet level label and then put at a shelf location. So, your WMS keeps up with what pallet contains what. Items that were manufactured on what line and what shelf location they live in. So then when the truck pulls up, that needs to get the pallet out to the end user or the customer. Then it tells the forklift operator, your WMS tells them, go to this pallet location, pick up this pallet and take it to this truck. The step that we filled in this process was to make sure that nothing else went wrong in the rest of that process, that when the operator went to that rack location that they pulled the right pallet and put it on the right truck. 
JT: Are there other metrics they're looking at when they're looking at the success of this system? 
CL: Yeah, I mean, there's a couple things. One, they look at, like you said, the verification of getting the right thing to the right place. I think what you said at the end there, Jim, is the right time. That's the other piece. They look at how many pallets are going through the door and how quickly that truck is getting loaded. And that was the biggest payback in this situation, was they were putting fifty percent more pallets through the doors. They were loading the truck fifty percent faster than they were before. and that's because the operator was able to focus on exactly what they needed to do, which was grab the pallet and put it on the truck. So that's the biggest metric they looked at for ROI. Anytime you're moving material, it really comes down to how quickly you can move it and how accurately you get the right thing in the right place. So that's the real driver in most logistics applications. 
JT: What were some of the challenges in placing manual inspectors with these things? 
CL: That comes down to the partner we use in this situation was Leo, and they were instrumental in making all the back end of the system work to make sure that the data from the reader got back into the tablet and then back through to the WMS. They went a step further, of course, and made sure that they could pull analytics and have a full visual picture on what the changes in efficiency were. But the reality is, we've got some tools built into our devices natively that make them easier to communicate with than another device. So if you've got a zebra handheld or a zebra tablet in place and you want to go to a zebra fish industrial scanner, then they already are married. So, you just basically pair them up and tell them what's there. And all of the back end you've already built to collect that data will take care of itself and put it into the WMS. So that's one of the reasons we're able to move so fast in this situation without disrupting anything, because we didn't have to reinvent the wheel. The physical installation is also designed to be streamlined and out of the way, and we took care of that as well. As far as the mounting of the dock door pillars and keeping them from obstructing the current flow of both forklift and operator traffic. So that's why we're able to do it so quickly. It's also something, I mean, Zebra's been doing barcodes for fifty years. This is not our first adventure into a warehouse. Right. So, we've got the processes refined. We've built our systems and software to make the implementation easy. So that way it saves time for the integrator and the end user. 
JT: How does one balance scan reliability and flexibility? When palettes and labels and operators all change shift to shift? 
CL: I think that there's a couple of things that go into that. Right. Obviously, it has to start with the hardware. You've got to have an image that's capable of dealing with variations and and depth of field and field of view. And that's where all of our fixed industrial scanners are built with flexible mounting options. Meaning you can auto adjust focus, you can adjust lighting. You can pair multiple scanners together. So, you can get a wider field of view, right. So from the hardware perspective, we've made that as easy as we can to install and varying environment. So, it's not like some other sensing applications and, and automation where you've got to meet a strict set of criteria. This we've designed to work with varying barcode sizes, varying barcode quality, varying barcode placement, lighting from the door. If the truck's not backed all the way up, we can deal with the ambient lighting that's coming through around the door in a daytime or nighttime situation. So we built all that into the hardware. I mentioned the software a minute ago. We can auto tune and adjust on the fly. We've got AI capability to help with decode and read rates. So we built the software to easily integrate back into the WMS or communicate with other devices. So all those things have been thought of and planned for to make the implementation as well as the long-term stability makes sense. And most installations. 
JT: And now this whole system, cloud based. I know you're communicating wirelessly, but is it going, is the data going somewhere to be stored. 
CL: In this situation, I can't remember exactly what Royal Canin is doing with their data. The majority of the time it's going to stay on prem. Most plants and most operations like to keep the data in the facility and on prem. So we like to remove the requirement for the cloud.
JT: It seems that one major driver is labor shortage. Nobody's cut labor to put the system in. However, do you see this type of system being kind of a long-term backbone?
CL: It is a long-term backbone. And if you look at it right, right now, labor shortage. Interesting. The environment that we're in and this is globally, this is not just in the US where we're facing a real labor shortage versus people trying to reduce the amount of labor on the floor. And that's what's driving robot integration now is labor shortage and training shortage or churn versus trying to save money, which is what we were doing, you know, fifteen, twenty years ago. So it really is a huge paradigm shift in trying to increase efficiency with the same number of people or increase efficiency when the people are constantly changing. So the days when you could hire a forklift operator and keep them for twenty years have pretty much expired. We've got a lot of churn in labor, which requires retraining. So even if you think about something as simple as this pallet level scanning before, you would have to train your operator on how to safely get off the forklift, how to look for surrounding traffic, how to find the right label to read on the pallet, how to make sure that label was uploaded appropriately. How to scan the dock door. There were a lot of steps that they had to do. So you get a new operator in. You lose days or weeks training that operator to do their job. Well, we did in this situation, what we do in a lot of industrial scanning situations is we just simplified the process. All we have to do is make sure your operator knows where to drive to get the pallet that they need to get and what truck to get it on, and if they go to the wrong place, you've now got systems and controls in place to tell you, hey, they went to the wrong place, take that pallet back off and put it somewhere else. So these checks and balances, we're not using AI inherently to drive the analysis, but it gives you that next level of visibility and the reduced need for complex or convoluted training for your operators. And this is just one situation where we're doing this. If you think about the inspection implications and the other fixed industrial scanning applications we deploy, they all really have the same impact of reducing the complexity of the operator's job and giving them more access to more information in real time. 
Outro: Well, that's a wrap for this episode of visions, produced by Endeavor Business Media, a division of endeavor B2B. Thanks very much for tuning in. If you enjoyed today's show, be sure to subscribe to the podcast and share this episode with a colleague who would find it helpful. Until our next episode, you can find us at vision dash systems dot com or on LinkedIn, Facebook, or for more insights, updates, and breaking news to keep you in the know. Thanks for tuning in. Until next time, stay focused on your visions.