LPRC - CrimeScience Episode 56 – The Auburn University RFID Lab ft. Justin Patton (Auburn University)

Episode Date: June 10, 2020

The RFID Lab at Auburn University is a research institute focusing on the business case and technical implementation of RFID and other emerging technologies in retail, aviation, supply chain, and manufacturing. Justin Patton, Director of the Auburn University RFID Lab, joins Dr. Read Hayes for a discussion on the lab’s origin, mission, past project highlights, and next steps.  The post CrimeScience Episode 56 – The Auburn University RFID Lab ft. Justin Patton (Auburn University) appeared first on Loss Prevention Research Council.

Hi, everyone. Welcome to Crime Science. In this podcast, we aim to explore the science of crime and the practical application of the science for loss prevention and asset protection practitioners, as well as other professionals. We would like to thank Bosch for making this episode possible. Take advantage of the advanced video capabilities offered by Bosch to help reduce your shrink risk. Integrate video recordings with point of sale data for visual verification of transactions and exception reporting. Use video analytics for immediate notification of important AP-related events and leverage analytics metadata for fast forensic searches for evidence and to improve merchandising and operations. Learn more about extending your video system beyond simple surveillance in zones one through four of LPRC's zones of influence by visiting Bosch online at boschsecurity.com. Welcome everybody to another episode of Crime Science the Podcast. Today I'm joined by Auburn's RFID Lab leader, Justin Patton. Justin and I have

had the opportunity to meet and chat before on the phone and we just did again. And here at LPRC, we felt it'd be a great idea to spend a few minutes with Justin on crime science and talk about the RFID lab, a little bit about its origin and history and the mission and objectives. What all the neat things that they've accomplished, how they work and operate, and what they're doing now and where they're headed. So, Justin, I want to welcome you to Crime Science today. Thank you. Sure. Thanks for having me. All right. Excellent. So, yeah. So, let's start real, you know, real quickly here at the beginning. Tell me a little bit about yourself and then we'll, and then maybe segue into the history, the origin of the RFID lab at Auburn? Sure. So I am an Arkansas native originally, and we started out at the University of Arkansas. So my background is physics was my undergrad and computer engineering was my master's. And it was

a weird, lucky, basically, we've had about 15 years of lucky coincidences at the lab and kind of kept us going and growing. But at the time, I was a master's student working on a project management course. And Dr. Hargrave was a faculty member in information systems. systems. And this was right around the time in 2004 when Walmart was very interested in RFID for case and pallet level identification and tracking through the supply chain. And I was his grad student. It was a weird mix because I came from the engineering school and he was in the business college. So I think it ended up working real well together because we had some technical expertise on one side and then he had the business case and the reason for doing it on the other side so that worked out well we um started the lab based on a project he

was doing walmart on the time on out of stocks with case level rfid and um that just snowballed so fast i had no idea what was happening in then because I was just a grad student. And I remember we grew from a single reader in the corner of a classroom to the basement of a building to a 20,000 square foot warehouse in the course of about four months. And at the time, you know, when we went to the full warehouse, Dr. Hargrave said something about, man, we're going to need a lab manager to run this place. And I was like, man, that sounds like an awesome job. I would love to have that when I graduated. And he's like, you want the job? And I went from, you know, poor broke grad student to a grad student with a job and then about 24 hours. And ever since then, the lab has just been kind of growing as fast as it can. So we've bounced from a lot of different topics over the years, but it was a lot of momentum in those days.

There was a lot of hype in those days. So it was opportunities for us to kind of figure out some of the teeth on how to communicate effectively with a developing technology environment, because there's so much snake oil and, you know, just nonsense out there, marketing, vaporware, being able to kind of cut through that and showcase, you know, what does work and what doesn't work and have a real conversation while maintaining enthusiasm. But, you know, not misleading people into what capabilities are. I know you have a lot of experience with that too. It's just one of those things where you want people to be excited, but you don't want them to feel like that.

They're going to get something that is not reality at the end of the day. Yeah. So, yeah. Yeah. And I, and just, you bring up a couple, a couple more than a couple of big points. But you and I reminisced for a minute here before the podcast recording. And I do remember working with Colin Peacock and Jerry and some of the team at Gillette. And they and P&G and many other suppliers were working at the MIT Auto ID Center back then, which was all about RFID and worked on a little white paper on the loss prevention side and got to visit. I remember

a smart shelf going into a Walmart up in the sort of that area of Boston and learning more and more about it, getting to go to the UK. the the point is hype and vaporware and what you brought up and i mean it was the topic to end all topics it seems like back then and the god whatever that was 2005 or somewhere in that range it seems like uh and it just i i think that that has provided some good but mostly not so good echoes all the way to this point, those that got involved and sucked in. And fortunately, in my opinion, my perspective from talking to people through the years, it really is Dr. Hargraves. It's you. It's the RFID lab that was there at Arkansas. And then we know where it is now. You'll talk about it here, hopefully. And that established, reestablished and have maintained the balance,

the credibility, the reality, as well as moving the technology and its effectiveness forward. So let me go back over to you. And so you're at Arkansas. We talked about, you've talked about, described the genesis and how you guys got rolling now, but you made a big move in the night. And kind of what happened around that and what's going on right now. Sure. But real quick, just to follow up on something. So were you involved in that initial Gillette RFID shelf monitoring program that they did? Yes. It was Smart Shelf. And at first, as you know, it was in a lab environment. Both Gillette had one in their Boston headquarters there

at the Prue in the Prudential Building is my recollection and seeing it there. And I think they had one too. I saw it actually over in Switzerland, but in their office there. And then they put it into, there was one in the MIT lab, and I can remember the building. And then they installed one into a Walmart store. And the read range, and this is, again, my recollection, was very short. And there was a couple out of, again, my recollection, out of New Hampshire somewhere that were anti-RFID.

And this was, you know, everybody would fly over New Hampshire somewhere that were anti-RFID. And this was, you know, everybody would fly over in airplanes and know everything you had in your cabinet, how many Charmin toilet paper rolls, things like that. And we were thinking it can't even read to the edge of a peg hook yet. There's no read range. You're not, nobody's going to be in an aircraft. So yes, Justin, that's my recollection. Well, and I'm glad you brought that because that project, the legacy of that project, especially when it came to a consumer notification and things like that have, they still reverberate. And we learned a lot from that in terms of, you know, just being transparent and not in the sense of negative or positive, but just how people perceive things, especially consumers and customers. And I'm sure you deal with this a lot with

LP measures in stores. You know, I've heard you talk about green shoppers and red shoppers and the different ways that they perceive their environment around them. But for the most part, people, in my experience, especially with RFID, if you try to be transparent with them about what's going on, then they tend to respond with ambivalence. Um, you know, there's been, I mean, literally in the, in the world around you, you're probably never more than about 20 or 30 feet from some type of an RFID tag in your daily life, whether it's in your vehicle for the parts as it was manufactured or in a retail store or dog or cat with the chip on them or in your phone or different areas and devices around you. But, you know, most people, as long as you, they feel like that there's some type of a notification or mark or something somewhere

on the product that they could look up at some point, they typically don't seem to notice that much. And another thing too, is like in 05, that group Caspian that was really, you know, anti RFID, they were against them. And usually they start out with supermarket cards. So like if you would go to a grocery store and they would give you the little barcode card that you would scan as a discount card. They didn't like those things. A lot of those privacy concerns that came out in 05 and 06, man, that was before Facebook. That was before Twitter. That was really before MySpace even. That was before we had an article in the news every six or eight months about a data breach with people losing millions of credit card numbers or private personal details and things like that.

private personal details and things like that. That was definitely before Amazon when you're used to being able to see recommendations based on your shopping list that pop up on the screen, no matter what you're doing. And I think a lot of the privacy stuff, especially on the retail side, has abated pretty significantly. But the stuff that tends to remain are the things that are a little bit more fanciful. So, you know, in a lot of ways, it's kind of a cool technology in the sense that it would be awesome if you could put an RFID tag on somebody and read them from a satellite from space, no matter where they are on the planet, from a spy movie perspective. But, you know, that's just not reality. But, you know, it doesn't matter how many times you tell somebody how it works.

The story that they tell, if it's cooler than the real story, that's going to be the one that sticks. So anyway, I'm kind of getting off topic, but I do think it's very interesting how the public perceives it because even now today, most people, when you say RFID, they don't know really what it is, or if they do, they have a very limited understanding, even if it's something they interact with on a constant basis. It's just one of those things that's kind of there, and it just kind of worked its way into our daily lives without really making a big splash. But since it kind of works behind the scenes, most of the time, it's not something that gets pointed out or noticed as much. No, I've even had retailers tell me that they, we have done census, you know, of the 60, 70 chains we work with and especially the apparel guys and the guys will be on it. We'll have a working group call and talk about product protection and supply chain and so

on. And someone will say, well, we don't have any RFID. The other guys will say, no, look, get an RFID gun or I'll lend you one. Aim that thing around your selling floor. You're going to be surprised how many products are already RFID tagged to your point. And it's funny because they're not all retail. A quick story. We went through a retail store.

This was last year. And just like you said, the first thing we'll do is we'll take a handheld through and just put some context around this. A lot of what we're talking about when we say RFID is passive UHF RFID, and those are passive tags. They're not battery powered. They're typically in a sticker format. They're a small flat tag. You've probably seen them on apparel items if you purchased apparel, especially basics in the last few years. But those tags work in about 950 megahertz range and you can use a handheld or an RFID gun to scan them. So we were going into the store to see if they had any tags around the area. And this store sold a variety of kind of indoor and outdoor goods, some appliances, and some big equipment

type stuff things as well. And we start picking up tags right away. So we use the finder, we go over to it and, oh, it's a printer that they had in here. And some of the companies that make printer ink put tags on the printer ink cartridges to track those students. That's interesting. So then we find another tag and we start zooming in on that and we go over to the area and, oh, it's a tag that goes on a, on a pallet level because right now for FedEx Freight, all of that stuff that goes through uses a UHF RFID tag at the freight level.

And there's one of those tags that's in the trash. It's like, that's interesting. So, oh, we found another tag. So we go over there and we find some cabinets, some wooden cabinets that were for sale. And a lot of the cabinet manufacturers use a passive UHF RFID embedded in the cabinet. So it helps them keep track of the different types and grains of the wood more easily.

And then finally, after all of this, we find some actual product that's RFID tagged in the store itself. And these guys are like, what in the world? Like, where is all this stuff? You have no idea. There's this whole world that's just kind of under the surface of identification for different items that we never see. of identification for different items that we never see. I mean, if you go out to a brand new car and you take a, an RFID gun out there, you're going to pick up, you know, four or five or six RFID tags, uh, that are used for different parts and components that go into the, uh, final assembly. So it's interesting. It's an interesting, you know, sub world of identification that doesn't exist really in a spectrum where we can see it but it's all around us so that's been that's really interesting and we go back your you know your

background in physics and computer science but one thing that i and this is really how i came to appreciate the role of physics in that not only with eas but what we talked about a little bit ago with privacy and so on. And this couple was really concerned and going on and putting out there about invasion of privacy by the satellites and aircraft and black vans rolling through the neighborhood. But it was all limited by physics. limited by physics. It seemed, you know, the amount, the read range, you know, and then the near field-ness of it, and then metal and water and things like that. So, Jessica, you take a quick, but a little, you know, tour through and describe to all of us that are not physics majors or certainly don't have near the depth and breadth of knowledge on RFID. What is it? And you mentioned passive and so on.

What are some of the types? Our main benefactor, Randy Dunn, which I know you know very well and work very closely with, of Centromatic fame, he's drawn up on a whiteboard and described to myself and our research team about the different types a little bit. But I'd love to hear a little bit about RFID. Yeah, okay. So, and if this gets too detailed, feel free to cut some of it out, but there's two main, first off, RFID stands for radio frequency identification. It is literally hundreds of different types of technologies. If you look in the history books, most of them will trace it back to World War II, where they used it for identification of friendly aircraft. Basically, they put a long resonator in some of the friendly aircraft.

So when the radar hit them, they would get back a certain signal. So the base would know that it was aircraft coming in that were friendly versus ones that may not be. And then that's evolved over the years. And RFID comes in two major flavors. Active RFID tags have a battery or power source on board and they'll transmit and they operate in a variety of frequency ranges. But when they transmit, you can beacon their location or you can ping it. You've seen these things going all the way back to the 1966 Batman TV show and every spy movie ever since. This is the idea of the tracker that you can put on something. It was a major plot point in Breaking Bad for one of their seasons. I mean,

this is the idea that we can put some type of transmitting device on something and then it will ping and tell us what it is or where it is at different points in time. And one of the most common applications of active RFID that people see every day is NFL. So in the NFL, every player, every game has two active RFID tags on a 108 shoulder pad. And when you see the post-play routes and everything, all that's RFID generated. So, you know, it used to be, you know, 20 years ago, they would show a replay and then you got John Madden just scribbling all over the screen in the yellow marker, which they still do some of that. But now when you see those routes and we see those after play reports and everything,

and even you see it live in plays when you see the separation between the players, when they're running down the field and everything, they have real time data on every player constantly in every game. And that's a massive amount of data to be able to analyze after the plays and things. And, and that same type of a setup, which is good for NFL players, is also good for people who are working in mines or on oil platforms from a safety perspective, because you definitely want someone to know where you are at all times in those scenarios. So there's a lot of different ways and reasons active RFID is used in manufacturing or whatever.

But active RFID tends to be slightly more expensive because they are battery powered, they're slightly larger, and they tend to have specialized use cases. They tend to be more closed loop. I think the things that most people have more familiarity with is the passive. And passive comes in several flavors as well. So I guess most people are most familiar with the pet chips that your dog or cat may have. It's like a little grain of rice that goes in their neck. And those are a lower frequency range. Those are HF RFID tags typically, and they'll give you information on what that cat or dog is, but it's very short range because water absorbs radio waves, metal reflects radio waves. That's

why we don't have radios on submarines. They use sonar because it's underwater and radio doesn't work great underwater. But a dog or cat body absorbs it. And that's why you can't just look on the internet whenever your dog or cat gets lost and find it, like find my iPhone. You have to have somebody that's close enough with an antenna to scan it. And you'll see HF and other things like the NFC payments for your phone systems and things like that. The most common type of RFID we work with now is UHF, as we mentioned earlier. And UHF is a little bit of a longer range technology. We can get, you know, 10 meter read range or so, depending on the type of reader. You can use handhelds,

you can use portal readers, you can use overhead setups, you name it. Stuff is everywhere. Every single thing that goes up to the ISS now on the space station that goes in those cargo transfer bags is RFID tagged and has been since 2013. There's about 10 billion plus RFID tags floating around the globe every year, mainly on retail products and mainly on a lot of apparel products. So about 10% of apparel retail currently is RFID tagged and that's growing at a fairly rapid rate. But the passive UHF RFID tag gives you a little bit of a longer read range. There's no battery on there. So theoretically they don't really go away or run out. And they typically carry a number on there called an EPC

code, which is a serialized data identifier on each product. So what they're doing is they're adding a serialized individualized identity to a product in addition to the UPC number. And that's where the real value is. It's not the magic. I can scan it from distance. I mean, it is nice to be able to take inventory very fast. But the real value in this is we're adding a serial ID to all those products that we didn't have before, which allows us to do a lot of things with, with track and trace and history that we couldn't do before. No, that's all good. A good primer. And, um, I understand, you know, the military has been a huge adopter. What, what all has the military done? You mentioned submarines and of course the role that water plays in absorbing or metal inflecting.

But what's the military doing with RFID? The DOD has done a lot of stuff, especially with active RFID, for tracking shipments and things like that. They use it at container level a lot of times. They've also done some things at an individual item level with item level with, um, well, you know, I think the, I can't remember which base it was. Um, but, um, there for, I want to say it was the air force Academy a few years ago was I've never been in the armed services, but apparently when you go in, especially before training, you get issued a lot of stuff. You get, you know, um, several different types of uniforms. You get a lot of apparel, you get a lot of things that you have to keep up with and you're supposed to turn back in later at some point in time.

So they were using RFID tags on all those garments and items so that it was much faster to check out and check in. And then also giving a serial identity to each of those items to make it faster to keep up with as well. And they've used it for some other things that are, you know, not as much on the public knowledge side with ordnance and ammunition and things like that. But the military has their own world and their own set of rules, and especially when it comes to identification, a lot of that stuff they don't like to, you know, broadcast.

But I think that the problems that they face for inventory control and logistics and and believe it or not you know theft I don't know if y'all have covered that in in your group but I mean there's a lot of stuff that gets stolen off the military along the way on supply chains and supply routes you know people tend to take advantage of the fact that there's a lot of expensive stuff moving through their area sometimes so they have some of the same issues everybody else does that need identification. Now, I mean, we do have the Army Air Force Exchange system as active members here at the LPRC, but not as you're talking about the actual movement of,

you know, what he's called beans, bullets, and band-aids and so on in their supply chain networks. So another thing that I've seen or I understand is growing, and that's the use of, say, medical supplies and healthcare facilities or on emergency vehicles to keeping track of what you've got to make sure you've got what you need exactly when and where you need it. Anything on that, Justin? Yeah, healthcare is a weird animal. Usually, we start with the business case, which is if this doesn't pay for itself, then people aren't going to keep doing it. And especially if you're going to be buying a lot of tags, you have to figure out what the cost of that is versus what the benefit and real dollar value is for the user on the other side of the fence.

So if you can't balance that equation before we start flapping tags on things, then we're just doing stuff for fun. And most people don't maintain expensive programs for fun for very long, especially in retail. So when it comes to health care, the issue we've always had problems with is understanding the ROI. The cost modeling is just wild, man. I mean, especially when you start tracking assets within hospitals or even items that are consumables, I mean, depending on whether, you know, what hospital or what area you're in, even a simple medication can be, you know, a dollar, it can be $50.

It is just crazy fluctuations in value in some of these things. And another thing that I found with healthcare is hospitals have a lot of RF stuff going on there. They got a lot of equipment. They got a lot of machines. They have just a lot of things in different areas. And we'll often find tracking systems, many different types of active or passive RFID tracking systems in these hospital environments. But they don't tend to be consolidated in terms of how they do their IT a lot of times. So you'll find, you know, the same vendor for an RF system on two different floors of a hospital, and those two different groups never even talked to each other, didn't even know that the same thing was in the same building. Whereas in retail and other environments, it tends to be a little bit more aligned.

So I think we've had a hard time figuring out, you know, what's the best way to approach these different environments. We had a hard time doing the cost justifications on those just because we don't understand the other side of the equation. But there is some momentum, especially on the drug side of the house. I know KitCheck has been doing a lot of good work with tracking the actual drugs going in and out of an area. And clearly, you know, we have more and more need for making sure that things aren't walking off. In today's environment, when you start talking about PPE and things like that, man, that's a whole nother can of worms. And I'm very curious as to what's happening there and how people can track and leverage some of those things. And I haven't had any RFID requests there yet because, honestly, I think people are just more concerned with getting that stuff in and pushing it out than they are and figuring out you know exactly how many and where they are

but um when at some point when this turns from just a pure push model into something a little more balanced i'm curious just to see how what the needs are for uh identification tracking in that world too no excellent yeah i i saw a We've got some, I don't want to leave anybody out. I mentioned Synthromatic and we've got Checkpoint. We've got Avery Dennison. We've got NeedApp. And I apologize if I've left any of our valuable members out, but we've got several of our solution partner members that offer different components or all the components of RFID. partner members that are offer different components or all the components of RFID. And we've been talking about, and that's what you and I've been conferring about,

is how do we support these retailers and others on the business cases? And I thought maybe, and you've brought that up a little bit, Justin, the ROI, the cost benefit. In other words, what is the business case for this? And that's been a big, to me, I've thought of you all in a few ways, but the technical side, let's make sure these the people that rely on Auburn University and your RFID lab can go and get the capital and the other operating funds they need to use RFID and leverage it to enhance their performance. So tell us a little bit, take a second or two here, if you would, on the business case. Absolutely. This is really the most important shift that we're seeing in the industry in general. And I'll say this, and I don't want to represent all the board members when I do, but we're an RFID lab, but I don't really care about RFID, right? I mean, RFID is great, but it's a means to an end. end. What we really care about and what's very important is adding a serialized identity to every instance of every item that we're dealing with in a store environment, especially in retail.

Because if we can track everything with its own unique serial identity, then we can move the entire supply chain and we can move the entire retail industry from this antiquated model of, you know, inventory accounting by SKU, quantity accounting, and we can move it to individual item track and trace. And that is way more data, clearly, because if I have, you know, 10,000 of an item, pairs of socks or something in an environment, it's easy to say this is the SKU number times 10,000. It's much more difficult to say, here are 10,000 individual serial IDs for that item. But this is what needs to happen. You know, we don't need to have the amount of waste in the supply chain that we do. I mean, you look at grocery environments now, we're throwing away

20 to 30% of the stuff that goes through just because of spoilage or just mismanagement or just generally stuff goes bad and gets lost or things get put out in the wrong order on the shelves and things like that. You've all been to an apparel store in your past where you go in and you look at it and you go to the section where your size or should be like a pair of jeans or something. And it's not there. The first thing you do is you look around and see how wrecked the rest of the store area looks. And you have to make a determination, is it worth me digging through the rest of this stuff to see if it's in the wrong location? Or is it worth asking anybody? Or am I just going to give up and go somewhere else? And that's a big

determining factor. And where a lot of consumers go is how nice and neat the environment is, or how organized it is, and how easy it is to find anything. And now this is way more important than it was a few years ago because now there's online options. There's pressures. You know, they're competing against some of the online retail channels for the physical stores. So it's not like the 80s where they put stuff out there and you go in there and if it's there, it's there. If it's not, you're out of luck. Now you've got other options. You can look on your phone right away. So being able to add that individualized item identity to all that inventory is the thing that is really helping. The first fundamental value of any RFID system is adding better inventory accuracy.

Most retailers roll at about 60% to 65% inventory accuracy by SKU. It's been validated in Europe, US, Asia, across the board, across various different types of retail. I'm not sure why it always hovers around that 55 to 65% range, but we found it over and over and over again. And so have others, universities and groups and things as well. And that's not acceptable. universities and groups and things as well. And that's not acceptable. I mean, when we're living in a world like today, where people are trying to order items online to go pick up the store, they don't want the store to tell them, sure, we got it, come down here and pick it up this afternoon. And then when you get there, you get a text message on the way down there and says,

nevermind, we couldn't find it. Or, you know, hey, you get there and oh, no, we don't have it, or something else, or it's the wrong size. Or even worse. You can't even order it. I mean, I hear this constantly and you probably experienced it on your own and everybody listening has as well as, you know, why is this store totally closed or why can some of the inventory I can buy online for pickup and others I can't like, why am I, why is it so hard for me to get these things that I know are right down the road from me? Why can't they send them to my house? Why do they have to ship them from a warehouse in Kentucky whenever I'm sitting right here three blocks from a store that I know carries it? So I think inventory accuracy is the number one fundamental thing.

Getting that inventory accuracy allows the retailer to do much more intelligent and things with more finesse and waste a lot less stuff. You don't have to buy as much stuff to sell the same amount. Well, the individual items are a key point, it sounds like, Justin. We know we've got some really neat technology from Digimark, for example, a digital or electronic watermark. We've got, of course, object recognition that gets better and better probably every day or week anyway via artificial intelligence, computer vision. So there are ways to recognize objects or brands or items, but you're describing this exact precise item number, this unit that exists within that type of item yeah and it and that

allows you to trace back the source you can look for ethical sourcing you can look for ethical supply chain uh shipping um things like that you can make sure that it's what it's supposed to be um so it's not counterfeit knockoff which is just that problem is getting out of control, especially in online marketplaces with fashion items. It allows you to do a better intelligent decisions about returns and things. And you're absolutely right. You know, computer vision, Digimark, whatever it is, if it's a QR code, 2D barcode, you're seeing some of this today. You know, a lot of the fashion shoe manufacturers will have a QR code in the tongue or inside

the shoe that a consumer can scan and they can verify that that item is authentic. Now, that's a serialized identity that goes along with that shoe. That can be cross-referenced against the RFID information that's on the box or on the outside of the shoe. And we can all refer those back to the same item. issue, and we can all refer those back to the same item. So we're moving very quickly into a world where each item has a digital identity, and there's multiple different carriers of that data that would get us back to that identity, whatever it is, however we do it, whether it's, you know, computer vision, or if it's just direct through RFID, or if it's a barcode, or whatever it may be. We expect to see environments where we can go through any of those channels for a positive

identification of an individual instance of something. And then, you know, the sky's the limit. The limiting factor is not the technology. And the limiting factor is not even the value proposition. The limiting factor is the accounting systems, because very few retailers, especially retailers at scale, are prepared to shift their entire inventory management structure to a serialized inventory accounting system and move right off of a quantity accounting system. That's a massive mega change to their fundamental inventory handling practices and software systems. And it's going to take a long time before most people are able to really catch up. And that's what I think is really going to start flying, I think. Probably like taking a sip out of a fire hose with that amount of data

to set it up, operate, and even just figuring out how to use it all, all the data. Tell me about, you don't have to name names, but we're both very well aware of, you know more than I, but the retailers that are pretty heavily invested in RFID for a lot of the reasons you're talking about. But a large department store chain that we both work with, where their merchants went out and decided that they would like to leverage this technology to really be better at omni-channel just in time and doing all the things that were important to them. What's a quick case study, Justin, for the retail use of RFID? Sure. A quick case study would be you would set up, typically we go a category at a time. You don't want to go out there and try to light up your whole store because that involves going back to your suppliers and get a lot of the stuff source tags. So you try to select a subset of suppliers. You get them to put tags on those things.

They start showing up in your store. It takes a little bit while for everything to flush through the supply chain. And then you can go out there and start cycle counting. And you can do that with handhelds or overheads or even any type of an exit portal or something like that. or even any type of an exit portal or something like that. But that is a method by which we, by cycle counting, we mean we take that serialized inventory and we compare it back to their existing store systems or existing on-hands. Usually, the vast majority of the time,

most retailers are overstated in their systems. You've experienced this. The system says we have one, I can't find it. Well, what happened to it? Well, it got stolen stolen or fell behind somewhere or some employee squirreled it away waiting for it to go on discount so they could buy it later or something, right? So there's various different things that cause distortion to that on-hand accuracy. What we're able to do with RFID is we're able to true that up and we're able to do that

more frequently. So it's not, you know, twice a year you're doing an annual barcode cycle count where you have people go through there and barcode scan that stuff where we're doing it on a daily or weekly basis a lot of times and getting those numbers better over time. So as soon as you get those better, the immediate result is your out of stocks go down. And then when you're out of stocks go down, your sales go up, because people find the things on the shelves that they want to buy. So that's usually the first impact. And then the second impact is you can do things like your buy online, pick up in store, reject rates start going down. So your customer satisfaction goes higher there. Also, you're selling more stuff to those online channels. And like you said,

you can do things like enable customer visibility to your total inventory in the store. So instead of putting a three item buffer on that store, so if it has three or less in the store, they don't tell the customer if they have any because they're afraid to disappoint them. They'll change that buffer down to zero so customers can see true visibility on that inventory before they even go to the store to get it. So that increases your sales through that channel as well. And then finally, you don't have to stock as much crap to sell the same amount, right? Because we're not having to just use this push model where we just throw so much stuff out there that the shelves are full and we hope people find it.

We're able to kind of fine tune and right size our inventory. So your sales are going up, but at the same time, your carrying costs are going down. And then you can do fancy things. I say fancy, more intelligent things like when it comes to loss prevention and asset protection, because then we're making determinations on what's going in and out of that store area versus what was actually purchased and what wasn't purchased. And I think that's where our worlds converge. Right. That takes some knowledge and it takes a little bit of data analysis and it takes a little bit more finesse than just going out there and saying this is how many of an item's in a store area. That's really where the rubber hits the road because instead of just busting bad guys at the exit, you're making more proactive and predictive loss prevention decisions, you know, either reworking the stores or finding some ways

to redirect poor behavior before it turns into an exit event that you're having to worry about. I guess, and I'm kind of curious about your thoughts on this too. Like, I don't know if it's turning the red shoppers into green shoppers or once they're incorrigible, that's permanent. I don't know, but you know, I think the idea is how do you make it more difficult or, or just, you know, How do you make it more difficult or dissuade them from trying to just walk out with things? I don't know.

Yeah, I mean, it is behavioral science there where we switch over, and that's hopefully our area. And that's what we spend most of our time doing, Justin, as we've discussed. of our time doing just as we've discussed, but we spend a lot of quality time with the people in those in situ, in those environments, the people that work there, the people that are shopping and moving through there, and of course, the people that are victimizing others there, the criminal offenders, you know, systematically observing them, talking with them, going off to the side or setting up scenarios and see how they respond to different. But really, it is marketing and how do we help them see, get in fear. But in other words, notice, recognize, and it'd be incredible enough for them to respond the way we want. Wait, if I take this thing out of here and don't pay for it,

then X, Y, and Z will happen. They don't know that it will, or they don't believe it will. They've got to know and believe it will, right? So that's the challenge that we're all trying to get right and dial in. And so whether it's EAS or another protective technology or it's RFID that's acting as EAS, really it's simple marketing that there's going to be a quick response and it'll be a negative response to what somebody does. So it's always stimulus response. If I do this, this is going to happen, will happen, and so on. And that continues. Can I ask you a question? So real quick short story, and I've always been curious about this.

So we had a system in a retail store one time. We had overhead set up so we could see the location and the inventory in the store in real time. And we were looking at it, and we were like, man, we see two video game units that are about, they're right over there by the exit. And that's not the video game section. That's kind of over in the apparel section by the exit of the store. So what the heck is going on? So these things are just acting weird. So we looked over there and sure enough, there's two young gentlemen standing over there in that section with a shopping cart that had two game systems and they're looking real nervous. So we thought, oh, okay, well, these guys are fixing the head for the exit. And so they waited and it took them about 30 minutes to work up the nerve for the exit to be cleared out. And by that point, the store had somebody waiting for them out front.

But I thought, man, if these goobers had just gone for the exit, they would have been gone. But they were trying to be smart, I guess. And they got over there and they got caught. Right. So we were able to kind of see some of that behavior before it happened but how often does that happen like is it different types of red shoppers but are most people like really good at it or most people just fumbling their way through or how does that how does that work in the real world well that's a good question and observation. So, first of all, the fact that they attempted the theft shows that we, obviously, the deterrence didn't work. Now we're talking about disruption, right? And documenting. Now we can, you provided, your technology is providing even more evidence about what in the world happened here. what in the world happened here. But what we've got to do is we're trying to figure out how do we deter them in the first place where they don't want to touch that half of them have to stick their hand into the fire, right? To see that it could burn. But most of them, you know, they're

not, they don't realize that they're going to, that that's a fire that's going to burn or they don't believe it will. So burn them, right? So that's, that's the big challenge. Now, the range of shoplifting, their intelligence, their skill, you can imagine ranges pretty broadly with most of not so good, mostly just lucky because there's not many people in the stores working there that have the time to watch and pay attention. They may not even be allowed or permitted to detain or pursue somebody that they know or believe has. So, you know, it's tough to gain deterrence. And that's what we work on our team every day is to do that. But that's what's going on. And so your technology helped understand, hey, we got a problem here or documented anyway, but it did not, in that case,

deter. So what can we do so this thing is something they don't want to touch is where we are. How do you do it up front besides wrapping a bunch of spider wrap on there and trying to scare them that way or something, I guess? Well, yeah. And the spider wrap is a good example where I mentioned see, get, fear. They're going to notice spider wrap. They're probably going to recognize it and understand they're going to get it. I understand what this is for. Now it's going to be a credibility thing, right? See, get, fear. It's going to be, all right, is that enough stimulus to get the response we want where they're like, all right, if I take this, somebody's going to know and catch me. If I try and remove it, it's going to make a bunch of racket and same thing.

So you've upped your game a little bit. And so like with EAS and RFID, we don't want it to be just totally invisible so that nobody even knows it's there. It'd be like slowing down speeders on a highway. If the police car is hidden, you don't even see the thing. You're not, nobody's going to slow down because of it. It's going to be, they of it um it's going to be they've got to be highly visible and credible and things like that so you know that's the same kind of challenge that we've got uh and so rfid though can have the dual benefit if we market it

properly maybe we can boost deterrence at the same time with the transparency that that or the the digimark or the you know computer vision all us, we got a better idea about what's going on, where it's happening, what doors it is going out. You mentioned going outside doors or other ways. They do that. We're always working on exits. But it's beyond a chess game. It's not checkers. That's for sure.

Because I've seen retailers that they want to hide the EAS pedestals, and then some of them want it out there. And I always thought if you hide it, what good is it doing it? Because people don't know it's there. And, you know, once you get out the exit, unless you just got somebody standing there in a mall, then they're gone, right? So I was always kind of, I don't know. I mean, I guess I always thought kind of half of hard tags was not the fact that it was a big

hard tag that was hard to take off. It's just the fact that you see that there's a big old hard tag on there. And we get a lot of research evidence that supports just what you were saying. It's yeah, they see get that and are probably a little more likely to be concerned about it. So you've got this competing interest here. What's very efficient to put on and things like that. Now, when it comes to, well, we don't want to offend our, you know, our shoppers. We

want, we don't want to do something to our aesthetics and so forth. We've got a lot of research evidence that shows most people are, most people just don't even pay attention anyway. We're zoned out as humans, right? Inattentional deficit is a fancy term, but people don't pay attention. We're looking at phones now all the time, staring at screens. So the idea that somebody's going to be offended by visible antenna or pedestals, our data show that has nothing to do with anything. In fact, most of them don't notice them one way or the other. The same thing with the hard tags. You're going to notice if you get home. I mean, you may notice it during the thing, but it's going to bother you when you get home and somebody didn't take it off. And now you've got to go from there. So that's the reality of it. Like you say, the data overwhelmingly show that

they've got to notice and recognize, and it's got to be credible, noticeable, recognizable, incredible to deter them or it's, it's not happening. That behavioral pieces. I just remember that scene in clerks from years ago, man, where they just had all the money sitting out on the counter with the change. And they had a little sign that says, you know, take, or, you know, leave what you owe. And they weren't even there. And the guy was like, well, how do you know that they're not just going to steal the money? It's like, if you just leave a bunch of money out on the counter like that, people are just going to assume that they're going to be watched and they're going to do what they're supposed to do. So I know that there's, you know, a whole behavioral component to it that is, uh, uh,

and yet you never can tell. And it's interesting the kind of way you divide people up to and by the, the different profiles, because it's hard to say who's paying attention to what that's in the store and everybody has their own motivations and reasons for being there too. So, yeah. Yeah, and it's tough. It's like an antibiotic, you know, you want it to be very specific to a certain bacteria to be more effective.

But you have broad spectrum antibiotics, right? They're probably not as effective against the individual, but they cover more. The same thing, we've got that issue. If we're just going after the professional, highly skilled offender, that's really difficult. They know to either avoid the place in the first place or they know how to overcome, adapt and overcome to defeat it. So you're normally in the middle. So you're normally in the middle. But particularly, so the one that's easily deterred is that normally the amateur person, like, oh, look, and they take something in the moment. But they're tough to deter because they don't notice and recognize things. Well, when we ask questions, what do you think that is over there?

Well, I think that is a radon detector up on the ceiling. Like, no, that's a security camera. So that's our big issue. That's why we go back to the see, get fear, notice, recognize, and respond model. And it gives us some dials to, as mad scientists, to work on to make things more effective and without running off the good guy. You know, do no harm is the first, but it's a little more difficult to do harm a lot of times because the good shoppers, we're just as clueless. The green shopper is just as clueless as the red as far as noticing and recognizing things.

Interesting. So let's go, I think from here, what we'll do is, where are you guys headed in the future? I think that for us, low-no-touch is critical to reducing the likelihood of infection from a pathogen like the COVID-19, but also for ease and convenience and other things that are out there. And then the idea of, you mentioned before, the specific item recognizability is critical because you can get the provenance and understand that. And then we call SCAG, reduce some of that counterfeiting and so on that can happen. And so what's in the future for the Auburn RFID Center and Justin Patton?

Sure. So two things on, and again, speaking to the retail front, because we've got a whole different work stream that goes on, works on aviation, aerospace, and all the problems they have there. But strictly in the retail world, there's two things. One is the category expansion. So we have a in the retail world, there's two things. One is category expansion. So we have a pretty good handle on RFID on apparel. We know how it works. People have deployed it many times. That's a path that's been trodden. So it's just on an expansion

mode for more and more items, more and more categories. We're looking at the new category types. So these tend to be items that are heavily researched online before they're purchased or they have high BOPUS scores. So it's things like sporting goods, electronics, those categories. And I think those are our next ones out of the gate that are going to be non-apparel. And that's going to help a lot in terms of just general market expansion. It's going to help a lot in terms of a perception from retailers. So they don't just look at it and say, oh, this is an apparel technology. This is truly something that is for all of my inventory. But I think our true passion, what we're trying to get back to is we're stuck in this awful fallacy that we've had since the beginning of RFID. I mean,

if you look up RFID and retail supply chain on the internet right now, you'll see all of these stories about how you can track and trace an item from the factory onto an overseas warehouse, and then into consolidation and shipping and customs, and then on to the US to their DCs, and then on through the full supply chain. This magical technology is going to help unlock all these efficiencies and benefits to the supply chain. The reality of the situation is putting RFID tags on these factories when something's made, when it comes right off the line and it gets a tag on it in Central America or Asia or wherever it may be, and it goes to the full supply chain and we never even scan or use that serial identity until we get to the retail store when it's just about to walk out and go home with a customer on the sales floor. And we miss all of those opportunities in between. There is a

river of RFID information or just serialized item information flowing through the supply chain, in the words of one of our colleagues. And we could dip into that at any point and use that serialized item data to make better decisions all through the supply chain. And that's where we're really trying to impact right now. The big problem is kind of what I mentioned earlier is not everybody has serialized item inventory accounting systems. So the chain breaks easily, especially when there's so many partners. We're trying to develop better systems or mechanisms for data exchange. I'll tell you right now, ASNs have not really been studied or updated or that technology for EDI hasn't been improved significantly since, you know, the 90s. And it is rife with errors. You'll talk to retailers and brands right now,

and they'll probably tell you, you know, their claims rates are one to two percent. Some of them have built-in claims costs in the orders to the retailers. That's insane when you look at the total volume and cost of the supply chain and you talk about shrinking things. Hell, a lot of the shrink happens because the item gets booked into the store inventory. It never existed to begin with because it was never even in the box, right? So there's a huge amount of opportunity to have some significant financial impact on the exchange of inventory between all those partners. There's a lot of efficiency that we can improve along there too. There's better ways to run a warehouse than taking a whole bunch of junk in and piling it on one side and then picking individual pieces out and pushing it back out the other side. We can make better,

more automated decisions. And you've seen this with packages of your own that you ordered through the mail. You think, why the hell did it go way up there before it came way back down here? You know, why is it going to five different locations in three cities before it comes back around to from the place where I ordered it to the place where I live? There's so many things that we can do much better and more efficiently. If we can learn how to use that information between all the partners in the chain, it's going to take a few years to get there. It's going to take a while to build confidence. But we're seeing that happen now, not just with retail, but moving into food. Restaurants are starting to pick that up with

their case level tracking. You see it in airports with baggage tracking. Delta, all their bags are RFID tracked as of year before last. And IATA has asked for all the airlines to do baggage tracking. So many items out there are learning how to use this kind of full supply chain visibility. And that's really going to be the next two or three or four years of driving towards more ubiquity for serialized item identification and where it's going to take RFID along with it. What a great wrap up, Justin. I really appreciate your time today. And for all our listeners, for more information on RFID, but and specifically the Auburn RFID lab, you know, it's rfid.auburn.edu. And you'll be able to track down Justin Patton from there as well, pretty readily on that site. So Justin,

I really want to, again, thank you for your time and your expertise and all that you've been doing to enable all these different enterprises and all these different organizations in the vertical markets to be more efficient and effective. And I see a lot of promise and you and I've talked about this and linking, tethering together multiple technologies from the watermarks and computer vision AI, but camera systems and others. And like you said, linking the chain together and doing a lot more here to get better, but doing it cost effectively. So to you and yours, stay safe and let's keep working together. And thank you again for your time.

All right. And thank you again to everybody for listening to another episode of Crime Science, the podcast here from Gainesville, Florida. Again, stay safe and reach out to us at LPResearch.org. And for Kevin Tran, our producer, thank you for tuning in to Crime Science. Thanks for listening to the Crime Science podcast presented by the Loss Prevention Research Council and sponsored by Bosch Security. If you enjoyed today's episode, you can find more Crime Science episodes and valuable information at LPResearch.org. Thank you.