How AI and Cameras Prevent Cold Chain Disasters

Pete Miller [00:37) Legacy tech is everywhere in logistics. Outdated cameras on highways, warehouse yards and security gates. Refrigerated trailers hauling vaccines. These old systems are costing companies millions in waste, energy, and liability.

But what if those security cameras could do more than just record footage? What if they could prevent spoiled food from leaving the warehouse? Or catch cargo theft before it happens?

Today on Predict & Prevent, I’m talking with Aymen Azim, co-founder and CEO of Technova Industries, and Jenna Faville-Azim, the company’s co-founder and Chief Marketing Officer. They’re transforming cold chain verification with AI.

We’ll explore how generative AI bridges the gap between legacy security and modern operations. Why visual verification beats traditional sensors. And how this tech is changing liability and risk for shippers and insurers.

From preventing spoiled vaccines from reaching pharmacies to filling visibility gaps during transit, Aymen and Jenna share their vision: A fully connected, automated logistics future. One checkpoint at a time.

Pete Miller [01:54]: So tell us how Technova got started? Like what problem were you originally trying to solve and how did you end up in the shipping and logistics industry?

Aymen Azim [02:03]: Technova got started when we discovered the huge problem in the security industry. Basically, we had a lot of legacy systems because this industry was there forever. And a lot of cameras, a lot of access control systems, and alarm systems were completely outdated.

So think about all the cameras on the highways, for example, in the U.S. A lot of those cameras were installed late 90s or beginning of 2000s, and it’s very hard to change. Because if you want to change one, you have to block the highway, get someone to climb a pole, get the camera, put a new one, rewire everything, and that costs a lot of money and interrupt operations. So basically, we thought about a technology that could connect to these legacy systems, old cameras, for example, and connect these old systems to the new technologies that we’ve been seeing today, like AI or even generative AI to be able not to have cameras just recorded, but have them manage operations and help with everyday activities. So that’s the beginning of our journey as an organization.

But we needed to find an application for a business case, a use case. And this is why we went into the logistics because it has that legacy system, and we’re trying to modernize it as well. So we ended up coming with a solution that comes from the security, but also apply to the logistics industry where we can solve a lot of problems and connect those old systems and legacy systems to the new technologies.

Pete Miller [03:42]: So as you were thinking about and getting involved, what kinds of costly mistakes happen in shipping and logistics that probably most people don’t realize? Like can you share some of the issues that your NovaControl solution is detecting?

Aymen Azim [03:56]: Well, let me then tell you the full story actually, because how we got in logistics, we found the technology, great technology, and we were at a security convention. And we were approached one with the largest, actually, largest grocery store in the U.S. We discovered that three weeks later when we visited the site and we noticed that in this warehouse, there is a thousand trucks going in and out every day in this warehouse. And we were so shocked by the manual processes that they had.

So a lot of problems happen because someone doesn’t check the temperature on the trailer or didn’t check the fuel gauge when the truck is leaving the facility. So this ended up causing a lot of problems where you have drivers sometimes going in, picking up the wrong trailer or a trailer not set at the right temperature, drive 500 miles either to deliver the wrong store or spoiled goods.

So this is where we found our technology very interesting, where we could connect to those legacy systems in this warehouse, connect to the legacy cameras at that same warehouse and make just checkpoints or checking the temperature, checking all the information, making sure the right driver is in the right truck, pulling the right trailer at the right temperature and enough fuel to go from the warehouse to the store.

And this was just moving from a fully manual process done by security guards at the dock or the gate to a fully automated process where we involve technology and legacy systems to complete an automation, a full automation process to be able to control all this information and make sure things are running smoothly. And that’s the power of the technology we developed.

Pete Miller [05:46]: So I was looking at your website. So can you really just sort of define cold chain verification? And can you give us an example how your solutions help prevent losses in that area?

Aymen Azim [06:01]: This is a really good question. A lot of cold chain verification was done through sensors that are inside the trailers and the trucks. We came up with a new way of doing it. We came up with a solution where instead of having to put sensors in each trailer, why don’t we use security cameras that are in the warehouses to check that temperature? Because every little trailer has a display on the driver’s side that shows that real time temperature and the set temperature.

So we just use those cameras, and, we can read the temperature through that display. So instead of having, for each trailer, a sensor, a hardware that adds a lot of cost to control the temperature in each trailer, we use one camera to sometimes control between 200 and 300 trailers a day. So it is a huge, a huge, I would say, modernization or just using those legacy systems to make them do other things than just record and, going back to the incident when once the incident happened already.

Pete Miller [07:18]: Can you describe for us how the technology actually works? So how does your AI technology actually see and what does it do when a truck arrives at the warehouse or leaves a facility?

Aymen Azim [07:31]: Generative AI and agentic AI brought a lot of modernization and a lot of other capabilities that the traditional AI didn’t have. Being able to read the temperature through visual information or visual images was something that was done forever. But with generative AI, this process gets from 90% or 85% of accuracy to 99.99% of accuracy. I don’t like to say 100%, but we are actually since we deployed our first system, we caught 100% of the errors going in and out the facilities we deployed the system at. I’m not going to look at it from the from the value of those cargoes we’re controlling, I’m just going to look at it from a human perspective. For example, when you’re dealing with vaccines or medical devices going from a warehouse to a pharmacy, the first thing you want to look at, how do I get this good at the right place at the right time and not spoil? Because if you’re going to put a vaccine for a kid, you can kill that kid. So we want to avoid that first before thinking about the value of the cargo. So I believe our technology is able to control that.

So when a truck is leaving a facility, we are able to look and build a checkpoint, a checkpoint that can ensure the integrity of the cold chain. And that checkpoint, we use that camera to verify the temperature in real time before anything goes outside or inside the warehouse. So think about, vaccines, think about medical devices, and think also about food.

It goes in our bodies at a certain point. Those foods, those vaccines, those medical devices, we use them on us. So we want to prevent anything that can that can go wrong. So we want to avoid problems that can occur to humans. This is why I’m looking at this from a human perspective first, and then I’m looking at it from a cargo value as well, because when you lose a full load of food, that’s a lot of waste. When you lose a full load of vaccines, that’s maybe people not being able to be vaccinated, or you’re shooting people with the wrong thing. So we want to ensure the integrity of the cold chain, and this is what we do when we have a truck or trailer going inside or outside the facility. We are looking at that temperature and making sure the right temperature is set on that trailer that is transporting the food you eat and the vaccines or medical devices you use.

Jenna Azim [10:22]: I think this really goes to our point that we consider ourselves a security company, and we see the cold chain and verifications in logistics as a security issue, not just necessarily operational. We’re using the security cameras not just to solve an operational problem, but to really, to Aymen’s point, to touch on a security problem of ensuring the security of the food, the vaccines, and the products that we ship that are temperature controlled across the globe.

Pete Miller [10:52]: So you mentioned that temperature verification has been around for a while. So what makes Technova’s solution unique and differentiates it from existing security cameras or other logistics software?

Jenna Azim [11:09]: Yeah, I will say that there’s a couple of reasons why what we’re doing today is a little bit different. The first I will say there’s a lot of data that comes from what we call telematics or sensors. And as Aymen was talking about earlier, you put a sensor in each trailer. You have data coming from each trailer. It’s very expensive. It’s not very reliable. And if you can imagine being the actual user, unless you have your own dedicated fleet and you don’t use anyone else’s trucks, it’s really complicated to get all of that data.

For example, you know, even when you have a large grocery retailer like one of our customers, they may have a dedicated fleet and they may have 900 trucks and they may still use other trucks that come in and out of their facility and they have no visibility into that. So we really are a checkpoint for a facility so that they can ensure that goods are coming in as they’re supposed to and leaving as they’re supposed to. So I think that’s one added benefit of cold chain verification.

It’s the liability aspect of it. You’re not accepting something that’s spoiled or under temperature. You’re not letting something leave that’s spoiled or under temperature.

Aymen Azim [12:20]: And to add to that, the telematics or hardware that sensors that are inside the trailers, gives you the actual temperature on the trailer, but doesn’t tell you the goods that are in that trailer. While with the camera, we’re able not only to catch the temperature, we’re able to catch also the trailer number, so we know what goods are inside that trailer, so we know what temperature should be there. And also the other thing with the camera, we can have a visual on the fuel level as well. Because if a refrigerated trailer goes off, you can it can be set at the right temperature.

But if you don’t have enough fuel to make that road, you may be able to reroute at a certain point when we detect that fuel level at a store or something like that. So we strongly believe that the visual it’s like humans. When we use our eyes to catch a lot of information, we have other way to catch other things, but we learn the word with our eyes. This is the same thing we’re using today. We’re given the systems, the logistics system, we’re giving them eyes through cameras.

This is what we’re doing today.

Pete Miller [13:44]: Yeah, I heard a statistic the other day, 40% of your brain just deals with vision.

Aymen Azim [13:38]: That is correct.

Pete Miller [13:41]: So the shipping industry involves a lot of different players like retailers and manufacturers and trucking companies and warehouse operators. So who are your target customers? And how does your solution give them a competitive edge in preventing losses?

Jenna Azim [13:57]: I would say our ultimate goal is really end-to-end cold chain verification. And what that means for us right now is approaching retailers, manufacturers, cold storage facilities and warehouses, people who have an interest in making sure that the goods that they’re storing and distributing or are incoming are coming in and leaving at the right temperature. So that would be, you know, the end user essentially is an ideal customer profile for us because they want to make sure that everything that they’re receiving is as it’s supposed to be. It’s the right load. It’s at the right temperature.

If it’s under temperature, you know, they know that, they know how to deal with that. And then when it’s leaving, a lot of times they’re sending it out to their own store. Or if it’s a cold storage facility, maybe it’s to a customer of theirs. So they want to make sure that, everything is as it’s supposed to be for their relationships and their benefit. So the competitive edge is kind of ensuring that everything that comes and goes is coming and going exactly as they say it’s going to and exactly as they think it’s going to.

And we’re just creating a virtual checkpoint for them on entry and on exit, so that they can hold up that end of the bargain for both themselves and their customers.

Aymen Azim [15:19]: I think the liability part of it is very interesting because our customers don’t want to waste food or any kind of goods that they are shipping. And from liability perspective, you want to avoid any kind of struggle or any kind of, anything blocking the supply chain from moving smoothly. So having checkpoints regularly, the way we’re doing it, is giving them a way to ensure that the goods are going the right direction with the right information and at the right parameters and configuration, however we call those. I think it’s very interesting to look at that from a shipper perspective and see, Hey, I have my goods on the road, and I am making sure they’re going the right direction, the right way to go to that consumer. Because at the end, they go to the consumer.

It is a very interesting process that needed a lot of manual intervention, and we believe we take that struggle out at least at the facilities as of today.

Pete Miller [16:28]: So what’s the role of the insurance company in all of this? Like, how do you see that relationship evolving?

Aymen Azim [16:35]: I think that was my liability point of last question, because one thing that we think about is how this play in the insurance world. Insurances understand risk, and this is very important. Then our goal is to reduce the risk for everyone, not only for shippers, but also for insurance companies. Because when you put a technology in place and you take your risk down, insurances are taking their risk downs too. And also from a cost perspective for our customers, for the shippers, they’re going to benefit from maybe the cost of the premium will be less if you have this technology because you’re reducing the risk as well.

And for insurance companies, they will have less claims. But at least when there is a claim, we know who’s liable because we have the technology to control that. We know where the liability stands. So we think that insurances are not going to be paying for things they’re not responsible for. So we have that piece of information in our software, and our technology can determine where the liability stand.

Pete Miller [17:41]: Yeah, that’s really powerful. I know you said that your first concern is to stop losses, right? So stop the vaccine that spoils or the food that spoils whatever. But do you have a sense of, I understand every case may be different, but what kind of cost savings would you be able to see?

Aymen Azim [18:06]: I think from our experience the industry for the last two years, we’ve been seeing two main goals for the for our customers to tackle. The first one is labor, and the second one is energy. From a labor perspective, and we know there is a big shortage in warehouse employees and warehouse people in general. So a lot of people want to reaffect those resources inside the warehouse instead of having them checking the yards and checking information about the trailers coming in and out of facility. So that is a big cost cut for our customers.

The second one is energy because when you have frozen food delivered somewhere at the wrong temperature, even if the froze frozen food are good at minus 10 degrees and you have to take them to minus 20 degrees, you have a big spike of energy. You’re spending a lot of energy to take that from minus 10 to minus 20. And most of the facilities in The U.S. don’t backcharge their customers that amount of energy that they spent because there is no way of controlling that. So having a technology that can tell you exactly what temperature those good goods arrived in a facility is powerful for, for these shippers and these cold storage managers to be able to say, hey, we have this amount of energy that we spend this month, and it’s because this shipper sent us these goods all the time under the temperature they should be at.

Jenna Azim [19:46]: Yeah, I would say it’s also, it’s an accountability issue as well. It’s holding, if this happens continually, you have the same load, you know, coming in continuously under temperature that you then have to or over temperature and you have to then lower. You then, you know, can hold that entity responsible saying like this continues to happen, this continues to be a cost and it’s not only bad for, you know, in that one instance. If you think about it over the course of time, it’s affecting the cold supply chain in general, like the loss of energy, the loss of food, and the errors that are occurring. And that’s ultimately what we’re trying to stop.

Pete Miller [20:33]: So where do you see this technology heading? Will human oversight still be necessary or are we moving towards fully automated shipping and logistics?

Jenna Azim [20:43]: I think what we always like to say about human oversight and our technology is that we’re not replacing these aspects of security that are true security. We’re taking the pieces that a human does that a human really is not the best at. Right? Humans are not great at data comparison and verification. So if you give a human two numbers very quickly and say in two seconds tell me if they’re the same.

You know, that’s not our greatest skill set. That’s what a computer should be used for. So we’re essentially giving that job to the computer. We give that job to the cameras to check and to AI to verify that yes, those two trailer numbers do in fact match. That’s the right trailer number that’s supposed to leave.

Instead, we leave the human the job of gut feelings and intuition. So you should still have a security guard at your facility checking to see if something’s not right. Is there suspicious behavior? Is there something happening that you can feel that you don’t think should be happening this way. Like this is what humans are good at using our intuition to solve problems and to suss out when something’s not quite right.

So we’re just, we’re reallocating that resource to free up more time for people to do that. It’s something that I think that part will never go away. That’s something that we’re always going to need humans for.

Aymen Azim [22:12]: I had a very interesting site visit recently, where I have this customer who did a lot of automation in their warehouse. And before moving to the full automation of their warehouse, like moving things inside the warehouse, they started by looking at the processes with humans, automated everything inside the warehouse, and then didn’t work and got more people involved to solve the problems the automated systems were doing. Well, the first thing that happened here, when I look at this problem, I look at how are we working today, how can we control the work today.

Today, at Technova, what we do is really putting in place checkpoints to look at how the supply chain is working, how the cold chain is working and making sure those goods are still moving the right way. We will get to a certain point where the system will be fully automated. But first, we have to put the guards to the system, the system guards at each checkpoint to be able to say, hey, with human involvement and system involvement, we’re getting better. And then see where we can improve things. I believe we can improve a lot of things.

We didn’t talk, for example, about cargo theft. When we catch those trailer numbers, we can read this USDOT number on the truck. We can also say, hey, this US DOT number, this company shut down last week. We shouldn’t see it on the road. We shouldn’t see it at the warehouse and trigger an alert to the security and say, hey, there’s something to look at here.

So we believe we can get to a fully automated system, but we go we have to go step by step, ensuring that the goods are moving the right way and then moving to, to the fully automated system where we can fill the huge gap in visibility, because there’s something we didn’t talk about. We talked about warehouses and stores and we can control temperature and if the goods are moving the right way, But we didn’t talk about that big gap in transit. When a truck goes from a warehouse to a store, a lot of things can happen in between. This is why we believe the next step for Technova is really filling that gap in visibility in transit as well. And we’re talking to a lot of gas stations where we can put our checkpoints.

So when a truck is going from a warehouse to a store and they stop to refuel at a truck stop, we’re able to look at that temperature again and say, hey, it’s still good. Or, hey, the temperature is dropping. There is a there is a store, actually two miles from here. Go to that store and drop all your goods there. So we’re avoiding that waste.

I believe we can add checkpoints in each step of this call chain until we cover the whole call chain visibility, and then we can go to that fully automated system, fully automated shipping system or logistics system.

Pete Miller [25:14]: So when you think about Technova in five years, that would be one thing, right? You could do sort of trucks in motion, if you will. But are there other big problems that you see that you hope to be able to address?

Aymen Azim [25:30]: I think one of them is just being able to centralize all the systems that are out there today. This is a big problem in the logistics industry. The systems are not connected. If Coca Cola is sending goods to Kroger, Kroger is not able to see what Coca Cola is sending them or controlling what’s coming on the road until the Coca Cola truck comes to the comes to the warehouse, and then they can check if the goods are good or not. So we believe there is a lot of problems that we can tackle there.

But also, we believe that the most important thing is to start by moving forward in the industry. We want to avoid the status quo, which has been here for the last 30 years. So that’s the big challenge we have today. And in five years, we believe this will change a lot because today the technology will allow us to do it. If we had this conversation five years ago, I would tell you we can’t do this. But today I can tell for sure that we can do it because it’s already happening.

Pete Miller [26:48]: What are the other considerations for insurance companies?

Aymen Azim [26:51]: There is this premium,  lower the premium for the users. I think it’s very important because with the risk reduction, because of that taking that risk down by a lot, I think it’s very important. We don’t know how to do it yet. But I think there is something there because they pay a lot of money for insurance and they send the claims, either they’re responsible or not, they send the claim to the insurance. If the insurance accepts it, they don’t care about the rest.

I just had a visit last week where every claim it’s sent to the insurances. And some of them are not their responsibility, but they still pay for them because they have no proof it’s not you. Well, I can explain it in detail, but it’s very interesting for insurances to know that their customer is the one liable for this.

Pete Miller [27:39]: You guys did a good job. Thank you.

Aymen Azim [27:43]: I’m excited to see it.

Jenna Azim [27:43]: Yeah.

Aymen Azim [27:44]: Or hear it.