Resilient Supply Chain — How Leaders Keep Business Moving
Resilient Supply Chain is for supply chain, operations, procurement and technology leaders who need practical ways to manage disruption, reduce risk and keep business moving.
First launched in 2019 as Digital Supply Chain and later known as Sustainable Supply Chain, the podcast continues today as Resilient Supply Chain, with the same extensive archive of more than 525 episodes covering supply chain technology, resilience, logistics, procurement and sustainability.
Each Monday, technology and sustainability analyst, keynote speaker and former SAP Global VP Tom Raftery speaks with the executives, operators, founders and innovators redesigning how goods, information and decisions move through global supply chains.
These are candid conversations about what works in practice—not polished PR narratives or vague predictions. Guests explain how they are responding to supplier failures, geopolitical shocks, volatile costs, capacity constraints, changing regulations and rising pressure to improve both performance and sustainability.
The podcast examines:
Faster planning and decision-making under uncertainty
Supplier risk, sourcing strategy and business continuity
AI, automation, visibility and operational intelligence
Warehousing, logistics and fulfilment performance
Scope 3 emissions, circularity and responsible sourcing
The systems, incentives and organisational changes needed to turn data into action
The central question is simple: what helps a supply chain withstand disruption, adapt quickly and continue serving the business?
New episodes are published every Monday at 7am CET. Resilient Supply Chain+ subscribers also receive bonus analysis, highlights and briefings on emerging industry trends.
Follow Resilient Supply Chain for practical lessons from the people solving real operational problems—and building supply chains that perform when conditions do not go to plan.
False alarms don’t just waste time. In cold chain operations, repeated unnecessary call-outs can teach people to ignore the next genuine emergency, leaving food or medicines at risk. Reliable alerts and clear escalation matter as much as detecting a fault.
In this episode I speak with Guy Yehiav, President of SmartSense by Digi. Guy recounts how a regional manager’s response to a 4am incident saved $50,000 worth of vaccines, illustrating why the human response matters.
Questions we cover
Can you trust a refrigeration controller? (08:15) Comparing controller readings with independent sensors can reveal discrepancies and help distinguish a blocked vent or faulty door seal from a potential cyberattack.
Why monitor shipments before they arrive? (10:33) With retrospective temperature loggers, medicines could already be heading to a hospital before an excursion is discovered; in-transit monitoring allows earlier intervention.
What should a monitoring business case include? (18:05) Alongside avoided spoilage and emergency maintenance, consider the value of redirecting staff time and the potential revenue lost when a stockout sends a shopper’s entire basket elsewhere.
If disruption hit tomorrow, would you know where your supply chain was most exposed? In 15 minutes my free scorecard helps you assess 27 resilience statements, calculate your score, and turn the result into three priorities and a 30 day action plan. You can download the scorecard free at tomraftery.com/scorecard.
Podcast supporters I'd like to sincerely thank this podcast's generous Subscribers:
Alicia Farag
Kieran Ognev
Gary Lynch
And remember you too can become a Resilient Supply Chain+ subscriber - it is really easy and hugely important as it will enable me to continue to create more excellent episodes like this one and give you access to bonus episodes of topical, timely supply chain resilience analysis.
🎤 Looking for a keynote speaker on supply chain resilience, AI, and sustainability? I help leaders understand the forces reshaping global supply chains, from geopolitical disruption and climate risk to AI and real-time decision-making.
Download my Speaker Pack: https://tinyurl.com/spkrpck
Finally If you have any comments/suggestions or questions for the podcast - feel free to just send me a direct message on LinkedIn, or send me a message using this link.
We rolled out a really large retail pharmaceutical in the US about a year ago, and they were using a different solution beforehand. They were using a solution that they developed. And their solution had a lot of false positives.
And you can imagine when you cry at 2:
00 AM in the morning, "Hey, there's a broken asset there, you need to go and fix it." And then you, you wake up, you wake your spouse, you go into the car, you drive to your store, you open it up and you see, oh, there was no issue, next time, what will happen, right?
Tom Raftery:
The cost of a false alarm goes well beyond the wasted journey. It can undermine the value of every subsequent alert. Good morning, good afternoon, or good evening, wherever you are in the world. Welcome to episode 149 of Resilient Supply Chain Stories and Strategies that Keep Business Moving. I'm your host, Tom Raftery. Today I'm joined by Guy Yehiav of SmartSense by Digi. We're looking at cold chains and a problem that extends well beyond them. The real value of monitoring technology isn't in how much information it collects, but rather whether it gives people the right information in time to do something about it. Guy, welcome to the podcast. Would you like to introduce yourself?
Guy Yehiav:
Absolutely, Tom. Thank you for having me. Happy to be here. I'm Guy Yehiav I'm the President of SmartSense. Entrepreneur by heart, like to solve problems with technology and innovation
Tom Raftery:
Okay. And for people who are unaware, Guy, can you give us a quick 101 on SmartSense?
Guy Yehiav:
So SmartSense by Digi is a company of Digi International, which is a publicly held company. We are working with industries that have compliance regulatory and need solutions for IoT, for sensing capabilities. Sensing for temperature, humidity, pressure differential. So main industries would be healthcare, retail pharma, pharmaceutical life sciences, and food manufacturer, food service, restaurants, groceries.
Tom Raftery:
And Guy, we've seen in the last number of weeks that hackers have been remotely changing refrigeration settings across European supermarkets. What should supply chain leaders take away from that?
Guy Yehiav:
Well, so let's talk a little bit before the cybersecurity attack and after. Even before the cybersecurity attack, the controllers that set those refrigerations, freezers, walk-in freezers, what we call assets, including actually reefer in the truck and transportation in transit, those controllers are set to temperature, regulated based on outputs where the air is coming actually out and making sure that they're actually working. The compressor is working. Some of them look for turbulences of the compressor. Others look at temperature of discharge. A lot of companies needed our solutions to monitor the actual product simulation temperature, right? If you have chickens, what the actual chicken temperature? If you have medicine, what is the actual medicine temperature? We have different simulators for a liquid, for example, glycol and others. And so the compressor and the controller will tell you that it's working, will tell you the discharge, but what if the gasket is being leaked, right? And what if the door is kept open? And what if there's other failures that are being done on the, the vent or over inventory? How many times you walk into even a grocery and you see that on the freezer, which is an open-end freezer, you have a lot of products being stacked above the freezer line. Well, guess what? The temperature and those products are not the same as the discharge on the bottom of the asset. And so we've been used to have what I would call first line of defence against failure, waste, or actually serving that, food. If you look now through the cybersecurity attacks that happened, bad actors were finding ways to get into the network or into the controllers. Some of those controllers are actually outside of the network. Other controllers are inside the network. Some of them are there to integrate with other areas. We can also tell you that the controller doesn't have power. We can tell you that the controller stopped vibrating. We can tell you that the controller doesn't discharge the right temperature or what the root cause of the temperature in the asset. But we are the first line of defence because we can tell you the root cause, open door, gasket replace, or, hey, someone is messing up with your controller temperature because we identify an excursion while your controller is saying that there's no excursion. And we obviously, we have, besides the root cause analysis that we're running, we're letting you know that there's an excursion using text messages, emails, phone calls and so the escalation is going through, and then you can just go analogue and take your controller out of the network, rerun it, and then we control your refrigeration So you can actually have real product under regulatory compliance temperature, so you can sell, continue to sell and serve your own clients. So we've been doing it all along, but now I believe that we get more calls because of the wave of that, cybersecurity effect.
Tom Raftery:
Hmm. Yeah. And are we making supply chains more resilient by connecting everything, or are we simply creating more ways for things to go wrong?
Guy Yehiav:
I think it's both, right? You always want everything to be connected because you would like to know and create knowledge. By the way, information is not the same as knowledge, right? So when you take all the temperature of all of your assets and you just say,"Hey, Tom here's your temperature." Well, what would you do with it? You can go and just feed it to ChatGPT, Claude or, anything like that and say,"Hey, tell me, tell me where I'm going wrong," but that's a one time transaction. Or you can actually create a more knowledgeable approach of, "Hey, based on historical excursions in the past, I believe you have a door open on aisle six at store 123, at a specific, you know, in Paris. So you do want it connected, you do want to feed it through a smart engine, but the critical thing is that once it's connected, you do need it secure because now once you have an opening for a bad actor, they can spur into your network, which is always the case. If you read any cybersecurity book, they say that the weakest link is a weakest entry point. Could be even just someone opening a laptop. I don't know if you heard about the latest cybersecurity attack on the you know, the Bitcoin but that was through an interview. right? So they got, an employee wanting to be interviewed. The CEO of that fictitious company ask him to download a test, and the test was infiltrated into the heist, largest heist in history of Bitcoin, right? So the weakest link link is-- could be a side, positioning when you're thinking about it, you would like all of the IoT to be closed loop and SOC 2 Type II compliant. and outside of your network, which is actually another, another point. The other point that I wanted to make is that when you build an IoT solution, the IoT solution should be a standalone that speaks without the solution when you want it, but should be isolated from your environment. And our solution is on cellular base, totally separated from your environment. Doesn't your use, your Wi-Fi, doesn't use your Ethernet, totally separate.
Tom Raftery:
But if, the equipment itself is compromised, how can operators know whether the information they're receiving is trustworthy?
Guy Yehiav:
Great, great question. And the way that you secure yourself from a first and second line of attack, if you will, is connect your controllers' information to your IoT environment. Have those sensors on top, so you have now two systems that feeding you information, right? And when we analyse an asset based on their size, we may put three sensors. We may put five sensors. So, for a walk-in, if it's a walk-in of a room like here, we may have six different sensors around, plus the discharge sensor of the manufacturer of the controller. We take all of this data in as an inbound, and then we compare it. And so if we see that the discharge is in one temperature and all six of the others, we could identify that there was a bad access. If we see that the six sensors position, different zone based on the vent, we will then identify which vent is being blocked. If next to the door is warmer, we can just tell you, "Hey, there's a gasket or there's an open door." Now, we also put a sensor for the door, so we know that it's open door or closed door. So therefore, we actually prescribe you a specific action. Now, if it's a what we call a catastrophic event, which means the out of power across the board, we do have power battery powered gateways that are good for 72 hours, so we can tell you the out-- power is out. The thing is that also we then continue to tell you which asset lost power, temperature degradated in the wrong direction, but it's not an emergency until we see that, hey, in about five hours, you're going to get spoilage. And now is the time to find a solution. Now, if we see that all of your stores are degradating in the same temperature in the wrong direction, we will tell you, "Hey, there's a controller issue and probably a virus or some bad actor that is doing something really bad across the board, and you should act on it now." So we already have what we call a catastrophic excursion type
Tom Raftery:
and where are the biggest blind spots in cold chains today, and why are they still there? Why have they persisted?
Guy Yehiav:
Let's look at it from a supply chain perspective. First of all the manufacturer, Looking at life science, food, ship it out to you. What you wanna make sure is that that shipment is not compromised as well. And a lot of time you have… The old way was to put a USB type data logger, we call this, you put it in the, transportation, and on the pallet. And then it goes out. When you receive it, you take that USB, you plug it into a computer, which is, by the way, a, bad actor could use that also against you. But you plug it in, and then at that time you can go and analyse the temperature. Of course, you're fitting it to an engine that will tell you if it was an excursion or not. The issue, by the time you do that, the pallet, if it's a life sciences medicine, could already be on the way to a hospital. Now you need to call and stop it because by the time you have a delay between the time you're receiving it, you wanna cross dock it and get it out as fast as you can through the warehouse, right? But, what if there was an excursion and there was a spoilage going on that you need to stop? So a lot of time you actually slot it into a specific slot in your refrigeration asset for a day until you can secure the fact that it was not spoiled, so you cannot cross dock it. What we then do is we build something that is called Voyage Solution. It includes that small sensor. It's a rechargeable sensor, and that sensor has both GPS, light, temperature, and movement capabilities. And so it identifies that it start moving. It knows where it is. If you load the route of the transportation for your transportation solution, it will then check that it is within the route. If there's a violation of the route within a certain mileage or kilometres, it will actually then alert you, "Hey, the truck or the pallet is going in the wrong direction." By the way, if there was a bad actor that committed any theft or any crime, then taking that off the truck, it will see that the truck going one way and this going another way. And then you actually for sure know that there was an issue. But then when it arrives to the warehouse, you know exactly that that pallet didn't have any excursion on the way. So now you can cross dock it and run it immediately to wherever, wherever you want. So that's one area of risk that you are running when you're getting manufacturing out. Now, you slot it in a warehouse. If you put it in the right slot. let's say that this, you know what, over the counter drug those gummy bears vitamins. If you slotted it up close to the roof in Texas, or in South of Spain, guess what? It will get 120, 130 degrees. And you need to be able to say, "Hey, that slot is not a slot that I can put OTC temperature." So we do have different consulting also work with warehouses on where to put different slots and actually tag them as, what type of product you can slot in which slot. When you take it out on, the road again- Put another sensor in, getting it to the store or to the hospital, and then, put in the right asset at the end game there to be picked up by a customer, by a doctor, by a nurse, in order to then secure it afterwards. So look at it from a supply chain perspective, there's an excursion that could happen you know, along the way both for regular type temperature needed as well as refrigerator or freezers.
Tom Raftery:
And let's say an alert comes through at two or three in the morning. What needs to happen next for that information to actually prevent a loss?
Guy Yehiav:
That's a great question. So we rolled out a really large retail pharmaceutical in the US about a year ago, and they were using a different solution beforehand. They were using a solution that they developed. And their solution had a lot of false positives.
And you can imagine when you cry at 2:
00 AM in the morning, "Hey, there's a broken asset there, you need to go and fix it." And then you, you wake up, you wake your spouse, you go into the, car, you drive to your store, you open it up and you see, oh, there was no issue, next time, what will happen, right? And so we just replaced that solution, and it was in North Carolina. And the story, 'cause I actually, got a call from the store manager. So, happened is that the guy that runs that store was on vacation in Mexico. He got the alert, and he was not aware, is the old alert or the new alert? I'm not going to do anything yet. Let's see what happened next. Within an hour, because he didn't accept the escalation, he didn't say, "I'm going to take care of it." He couldn't take care of it. He was in Mexican on vacation. His manager, who-- she, she, is the director of the region, got it and said, "You know what? We just put a new system. Let's prove it wrong or prove it right. Let's actually go and do it myself." She got in the car, she drove an hour and a half to get to that store. It Was not a close store for her for her.
She went at 4:
00 AM and she saved $50,000 of vaccine because there was an excursion. And so the next day she actually called me and says, "Hey, we just created a first ROI of a saving in the middle of the night." And it's so critical to suspend any false positive as much as you can without losing inventory. And so we have-- today where you have so much data available that you can use in order to identify is it a false positive by what percentile probability. And so we actually use that to suppress those false positive and only escalate it when is a true positive where you can still save, the, the inventory.
Tom Raftery:
At scale, the scarce resource isn't the sensor, it's the attention of the person who's expected to act on it. Every unnecessary call out spends some of that attention, and that's a resource you cannot afford to exhaust. Are there things that work in an initial deployment that become harder when scaled across hundreds or thousands of locations? I mean, there's gotta be right?
Guy Yehiav:
Obviously, what-- when we get a new a new installation or a new customer, we actually do what we call an initial rollout. We start it. We have two weeks of stabilisation process that we-- the system checks itself. Today, with a machine learning and the AI model, it can actually check itself. You have human in the loop, of course, to check it on top, because you don't want to hand it over a system that doesn't work. But it takes two weeks stabilisation and it's ready to go. And we implemented that 2,700 stores in less than nine weeks. So you can imagine getting to all of it in the US, the US is pretty wide, getting it to all the stores, installing it, train it, and then stabilising it. But the escalation is very clear, and so, it's fairly easy.
Tom Raftery:
And when does continuous monitoring genuinely pay for itself, once you include deployment, integration, maintenance and the people needed to respond?'Cause there's a lot of other extra costs.
Guy Yehiav:
Yeah. and so we, we actually have a, an Excel spreadsheet that takes you through all of that beforehand. So you have also a lot of soft that is not easy to calculate. I'll give you some examples. Number one, if you are actually going to test temperature manually today, you have a person that goes and do it. Well, that's first of all a lot of frustration because no one-- who wants to go in check temperature and put it on a log manually? And so there's a lot of soft dollars that you can find. The time that it takes to do it. Well, you're not going to fire those people, but you can reroute those times to serve your customers both in a warehouse, on a truck, or in a grocery or retail pharma type stores, right? So there's a lot of soft dollars that you can count. on. We have a lot of-- We have about 12 entries of those. The hard dollars is, well, you're not going to lose as much, you're not going to spoil as much, and you're going to call the technician not in an expedite manner. So if you look at, hey, how many times did I call a technician in the last 48 months? Calculate that. you are going to save that dollars because now, before and after this failing, we will tell you that gasket needs to be replaced. We'll tell you that freon is is, leaking immediately when it start leaking. We will tell you that a door is not closed here, the, the right way with the vacuum because we'll identify the root cause. Now, you can actually set the, technician, "Hey, when you are in the neighbourhood, can you stop at that store? Can you stop at that warehouse? And by the way, this is what I want you to do." And now, when it's prescribed and it's before, the asset will not fail, so you don't have catastrophic, you're not losing waste, and you're not expediting those technicians and paying even extra money for it. And you're not losing by not having inventory being sold that you could sell and, and serve. We have a, customer actually Lowes, Lowes Food in North Carolina. They have about 170 stores. So med banker presented at our live event just about three weeks ago, and he was talking about the fact that how they calculated the ROI before and how they calculate after. And what they identify is that overall revenue and service went up because we don't have a failure. Can you imagine going on Thanksgiving or Christmas Eve to buy your meat patty for burgers and hey, there's no ground beef. Why? Because the assets fail. Well, what do you do? You drop your basket, whatever you actually pile, and you go to another account, another grocer to get your beef patty. But you just lost, not just the beef patty that was spoiled and rotten, but you actually you lost basket, whole basket being lost. You lost that, loyalty of that customer because the customer could not-- maybe not even come back later, right?
Tom Raftery:
And what looks perfectly sensible in a supply chain strategy presentation, but consistently fails in operations?
Guy Yehiav:
I think that's the human factor, right? The human factor. A lot of false positive. A lot of, when you have human in the loop, or when you have a human actually going and checking the temperature, there's what we call pencil whipping in the UK and US. I don't know how to translate it to European, but basically it says, "Oh, I need to go and check the temperature. I'll just walk through, and I look at the analogue, dial, and if it's in green, I'll just write 30 degrees, 32 degrees, 27 degrees, and I will not actually check it." And, That could create spoilage without you even knowing.
Tom Raftery:
And we're moving towards being able to monitor almost any physical asset. What, what becomes possible when that capability is available throughout the supply chain?
Guy Yehiav:
So that's a great question. So I call it sensor ability extension, right? And we always wanted a SmartSense. We looked at sensor ability extension beyond just temperature, humidity, and pressure. We wanted to identify occupancy, so when the cashier is next to the cashier, right? Or where, how many people are actually sitting at a desk. We wanted to do proximity. So if you have a fire distinguisher in the warehouse or in the store, you put a sensor next to it and say, okay, when-- if someone takes the fire extinguisher off, I want Tom to know that that specific extinguisher number nine at a store 987 was taking out, and you need to replenish that, right? And so we wanted more and more sensors. And so the good news we just announced not so long ago, that we are acquiring a really nice European company out of Oslo, Norway, called Disruptive Technologies. In an IoT solution- the battery life of that IoT sensor is so critical. SmartSense was able to optimise it to about five years. So every five years you need to change the battery. But Disruptive Technology were able to do it for more than 15 years. And so now we can sense a lot of different things that we couldn't sense before. Like CO2, pressure, gases, uh, water, leak, right? So we can identify that the asset is not just bad on the inside, but it's also water leaking outside. Or maybe inside it's still okay, but the water leak on the outside means that it's going to break later, right? So we are going to take all of these data in. And when think about the physical world, all of these data that you send, you cannot download it. You cannot ask ChatGPT or Claude to make it happen. You need to actually sense it. In order to sense it, you need someone physical in the physical world, that will then it into a digital world, load it, learn it, understand it, prescribe action back into the physical world, so human in the loop can go and act on it.
Tom Raftery:
And where do you think the industry is heading in the wrong direction and where have you changed your own mind?
Guy Yehiav:
I looked at my own house and my friend house at the time, five years ago, And you have a lot of sensors in the house. You have a temperature, you have humidity sometimes, you have shades coming down. You have Sonos for music, or Bose or Sony. And so you have so many, but think about it for one minute. On your app, how many apps do you have running for it? Well, you have Sonos for music, you have one for shade, you have one for temperature, humidity. So fragmented. Well, in the business, you cannot be fragmented. It needs to be one application and one backbone that can analyse all of this data to tell you the real issue. I'll take an simple, conference room. Conference room with the shade up, sun is coming out, it's very hot. But if there's no one in, I don't need the HVAC yet because later on, I can take the shades down, and I can cool the area and save money, energy. So I need to know how many people are in their office. Are they sitting at the desk of the conference, or at their own desk? Is the shades down? What's the temperature and humidity, right? So, all of that, having it as one IoT platform will be able to help you manage it. And so the, the, idea was that this industry is so fragmented, we need someone to be the IoT backbone.
Tom Raftery:
There's an important distinction here. A shared view of the data doesn't require every device to share the same network. In fact, the cybersecurity example earlier is a good argument for keeping those two things separate. And looking three, five years down the road, what do you think will distinguish genuinely resilient cold chains from everyone else?
Guy Yehiav:
I, I think that because cold chain demanded better IoT and better connected solutions across the board it will be a one system across handshake. I was speaking with another 3PL, CEO, on the West Coast in Arizona the other day, and she was talking about the handshakes that right now it's very integrated, handshake of what was on the BOL? What was on the truck? Where did I get it from? What's the temperature? How long did I drive? What was the excursion on the way? Handshake into a slot into the new warehouse. All of that right now is, hey, I need to transact it in a way of integration. Once all of that will be, let's say on a blockchain or an handshake, that you can actually then create same ledger, same dictionary that will move me at the grocery chain picking up a finish. So having that visibility will be, will be great. Another thing about vaccine. You go and get a flu vaccine, who tells you that there was an excursion or wasn't an excursion? You want to be able to read the vaccine specific and see that there was a handshake, there were handshakes along the way, and there was no excursion of that vaccine as well, right? That's where we are going. And, cold chain, I think, is on forefront of that travel journey of integrating that data across the board.
Tom Raftery:
All right. And and for supply chain executives listening to this, what's the first thing you'd recommend that they do differently tomorrow?
Guy Yehiav:
Look at the data from a perspective of, hey, I cannot download the data, therefore, I do need sensors. Doesn't matter where you buy them. Consume that in and start training their language there, whatever engine they use, on what is the root cause of any failure based on the data that they log in. Even if they do it with latency, better than not doing it at all, right? So then you reduce-- start reducing the latency.
Tom Raftery:
Alright, and, and a left field. Question for you now, Guy, if you could have any person or character, alive or dead, real or fictional as a, as a champion for building genuinely resilient supply chains, who would it be and why?
Guy Yehiav:
Well, my a, a, a good friend of mine who is a well-known professor at MIT, Professor David Simchi-Levi, he and I built some formulas for re supply chain resiliency. And always fun to speak with him, always. So if anyone have a chance, or if he's presenting in any school next to you, please go and visit him and send regards for me. His resiliency bible, he puts things just that are very complex in a very simple way. And so if you see any one of his presentations, he's talking about from a resiliency perspective, getting all the variable and looking at time, to survive and time to fix. And so obviously, time to survive needs to be longer than the time to fix because that means that you still have time to to fix whatever on hand information, waste reduction or anything that you need to do. But resiliency is, resiliency is number one on any agenda for supply chain should be. And if you're looking at it from a culture perspective it's aligned as well. Yes.
Tom Raftery:
Guy, if people would like to know more about yourself or any of the things we talked about on the podcast today, where would you have me direct them?
Guy Yehiav:
About myself, connect me with and LinkedIn. I always intrigued by smart people challenging us and talking about interesting topics. There's a lot of them today and anything but politics, right? and then, uh, SmartSense.co
Tom Raftery:
Okay. Superb Guy. That's been really interesting. Thanks a million for coming on the podcast today.
Guy Yehiav:
Thank you for having me, Tom. I'm looking forward for the next time.