Save on Cold Chain Costs: Reduce Transportation Spend and Improve Service this Winter

 

Hello everyone and thank you for joining us today for today’s webinar, Save on Cold Chain Costs: Reduce Transportation Spend and Improve Service This Winter presented by Everstream Analytics. My name is Lauren McKinley and I am joined today by my colleagues Jon Davis, Chief meteorologist at Everstream, and David Shillingford, Chief Strategy Officer at Everstream Analytics.

For today’s session, this will be recorded. All attendee lines are on mute. If you have any questions throughout the presentation, please add them in the questions box in the go-to webinar panel. We will make sure to address as many of these questions as we can at the conclusion of the session. And with that, I will turn it over to Jon.

Jon Davis:

Great, thank you so much Lauren, and good day everybody. So our starting point in this discussion today, we’ll be looking at winter conditions and how to optimize that from a transportation or cold chain overall standpoint. But we want to go back a little further and talk about the baseline. And the baseline and the first thing to talk about is really extreme weather.

So if we kind of look at extreme weather here overall and talk about the winter, and this is going to be certainly discussion where we talk about overall temperatures, certainly lately we’ve had a lot of extreme weather. Certainly at the top of mind the last couple of weeks is some of the hurricane activity that we’ve seen, Hurricane Ian here, and the track of Ian that made its landfalls in Western Cuba and of course southwest Florida, and then finally the Carolinas, and some of the issues that cause within the supply chain network overall. And prior to that it was Hurricane Fiona that did major issues within the supply chain and caused disruptions in Puerto Rico and the southeast portion of Canada.

These kind of extreme weather events, it’s nothing new. And we’ve seen for a long period of time that overall extreme weather events have becoming more frequent with time. So when we look at overall extreme weather events, and this graph here is looking at the billion-dollar disaster events just for the US, there’s been an increase in these severe weather events in virtually every category if we look back at the last two, three, and four decades. In other words, extreme events are becoming more frequent and it’s certainly a trend that we’ve seen. And that trend has been increasing even more so over the last five or eight years, again in virtually every category out there. And it’s these really extreme events, the tails of the distribution, that really cause the biggest issues within the supply chain here overall.

And overall, now we’re going to talk about how does this relate to winter, and how does this relate to winter temperatures? We’ve also seen this weather volatility in winters over the last five, eight, and 10 years on both sides of the spectrum. In other words, we’re seeing more extremes in temperature patterns across not only North America but all sections of the globe, Europe, East Asia. And these kind of extremes that we’re seeing have certainly been more evident over the past five or eight years than prior timeframes overall.

This image here that we’re looking at is just a general image. Looking back a couple of years ago in Feb of ’21. And the big Texas polar vortex freeze had major ramifications across many areas of the US and Canada. And some of those ramifications reverberated on a global basis and still actually some impacts even today from that. So that is one direction that we’ve seen, is that we tend to have more warm winters, but when we tend to get cold it tends to be extreme. Again, the tails of the distribution overall. On the other side of the spectrum, we look at the warm side of things. And again, there’s been more periods of warmth across the winter timeframe, especially across the US and southern Canada over the last five years compared to the prior timeframes behind that.

But we’re seeing evidence that you tend to get extremes on both sides of things. And to us that really means opportunity. So when we think about extremes and opportunities in deciding overall transportation costs and transportation decisions out there, during periods of warmth, that can change much more than it was let’s say 10 years ago overall. And as we look ahead to this winter, and again our official winter forecast is coming out in a couple of weeks, but there are a couple of drivers that are in place right now that we think will be very important in the kind of temperature trends and storm systems that we tend to get across the US and other areas of the world as we go toward winter here coming up.

And those macro drivers, number one is the La Nina event across the Equatorial Pacific. The overall period of colder than normal waters in the central and eastern sections of the Pacific Ocean. That changes jet stream patterns, especially downstream in a place like North America. And that can have alterations and certainly tendencies that you tend to get with those types of events here overall. We also have one of the other drivers of course, the polar vortex, which every winter is extremely important here overall.

So as we begin to kind of look ahead to this winter, the fact that we have a La Nina event, and that tends to disrupt or change Jet streams patterns, and we think this La Nina event which has already been in place for the last two and a half years, will be in place through the end of this year and into the early portion of 2023. We think that will tend to be and have major ramifications on the overall pattern across North America. Typically during La Nina winters, we tend to have increased extremes, increased volatility across areas of the US and the southern portion of Canada. So with that in place, we expect a highly volatile overall winner pattern coming up across much of the US and many areas of North America overall. Typically during La Nina winters, you tend to have more extremes, many times on both ends of the spectrum.

And again, this means opportunity out there and certainly cost savings here overall coming up across those areas. And again, from a standpoint of the cold chain as to how things used to be, you go back 10 years ago, it is not business as usual. These extremes that we’re seeing not only in North America but on a global basis, to us really means opportunity out there. And that’s one of the issues that we’re going to talk about next year coming up is some of those opportunities, some of the solutions out there, here overall. So I’d like to pass it to my colleague Dave Shillingford, and he’ll talk about some of these opportunities. David?

Dave Shillingford:

Yeah, very good. Thank you very much, Jon. Okay, very good. Well I think that’s a great summary of the world that we now live in. And more and more how weather is having an impact on everything that we do. And just to think about one of the lessons that we learned from the pandemic, and that is where you have an external risk, and an external risk that essentially is impacting everyone across an industry, it’s as much an opportunity as it is a risk because it’s impacting everyone. And so the companies that are better prepared and the companies that can better respond to whatever it is that comes, are going to have a competitive advantage. So I’ll loop back to this at the end of my comments, but the idea that risk is always a bad thing isn’t really the case. Companies who think of risk as an opportunity are the companies that are going to have the best chance of gaining a competitive advantage when the weather is worse than expected or even better than expected.

So for the next 10 minutes, we’re going to talk through this top down. We’ll start with network design, we’ll talk about network planning, then we’ll get into shipment planning and some of the predictive analytics that companies are using to really turn this type of risk into predictive prescriptive analytics that give them a competitive advantage.

But before we dive in, I just want to talk briefly about… You may have heard this phrase, analytics as a service. And it’s important to define what I mean by that. So if you go to the next slide, there’s five components of this that you should be looking for, because it’s the kind of thing that’s going to become or already is a buzz phrase. But it has a really practical importance. And that is the analytics that are being delivered are being delivered out of the box. This is not a month, year, two year process until you get your first insights.

The analytics that are immediately adding value are delivered out of the box. And that means that whoever’s delivering these analytics needs to have a number of different things. One obviously is data scientists with the right skills. Second critically, is they need to have data, the right type of data, and also to have subject matter expertise and secure and scalable platforms, because there’s an incredible amount of data involved in doing these things. So you’ll see a couple of the examples I give here. We would describe as analytics as a service, and this is how we think of it. But it is, if nothing else, it’s a big investment.

So when thinking about design, network design and network planning, the way we think about this, at least from an environmental risk standpoint and other risks as well, is that for network design you’ve got to look as far out as you can and that you need to. And some of the decisions you’re making that the location or the building of facilities are multi-year decisions that are being made. So that’s where we start thinking about analytics that would be described as climate analytics. These are long term forecasts that we are making that relate to things like water availability, or heat and humidity stress, wildfire risk, flood, sea level rise, things like that that need to become part of network design going forward.

In the medium term, we’re thinking more about seasonal planning. Jon just mentioned that we’re about to release our seasonal forecast for this upcoming winter and that is used to enable companies to think about their network planning, not necessarily to change anything, but at least to have plans in place, knowing where the risks are likely to be and what those risks are likely to be.

And then you come into the near-term planning where we’re talking more about forecasting and where the risk scores for certain nodes, facilities, or the lanes in between them are showing up numerically, and that can start to drive business decisions. An example of this, although it is an extreme one it’s a good one, was the polar vortex that caused widespread freezing in Texas, February 2021. And in that case, our clients were already thinking about this because of the seasonal forecasts that they’d received prior to the beginning of the year. And then as that became a forecast and the risk scores started coming into the platform, they were actually starting to make decisions. Some of them were bigger decisions than others, some took longer than others, not moving temperature sensitive goods into the area. In some cases, they were moving goods out or moving them to better protective facilities. Some of our clients actually closed down production facilities at certain points. And they saved millions and millions of dollars in doing so but could never have done so without the warning and without the confidence in the accuracy of the analytics that we were providing to them.

So how can this type of data and these analytics be used in shipment planning? So, it’s kind of a busy slide, but the top half you already know. This is what you do as logistics professionals. I’m going to focus more on the bottom half where I’m talking about the type of inputs and ultimately insights that are being input into transportation planning and execution systems. So, I’ll focus less on the in transit and less on the lead time analytics because I think that the shipment planning analytics is the really important thing.

This isn’t just weather risk. If we’re thinking about the risk to on time delivery, weather’s a very important factor, but there are other operational risks. There are other external risks that need to be taken into account. So, at a high level, we’re looking at historical shipment records. Millions and millions of historical shipment records are being fed into these machine learning models. Same is true with the predictive weather forecasts, because what happened yesterday might not be what’s going to happen tomorrow. We need to have predictive risk as part of these models.

And not just weather, there are other types of risks that we’re analyzing that go into these models. And the machine learning models are looking at over 60 variables, and they’re learning all of the time so that they get better and better. And this means that planners can look at the shipments that they need to so that they’re being more effective, more efficient. The models actually throw up the variables that are having the biggest impact to on time risk. And so that risk score that is shown against each shipment can then be used to make adjustments and focus on the right shipments and look at the variables that can be changed if they can. And if they can’t be changed, at least this is providing early warning to the downstream recipient of the shipment. And just to emphasize, this is happening weeks and days before the shipment leaves. This is not something that’s… It can be done in transit, but this is being done during the planning phase. So that’s analytics as a service as it relates to on time risk.

I’m going to switch now to in full risk. And this is bringing it right down to a single risk and that is temperature. And Jon said at the top of the call that temperature volatility is one of the things that we are seeing more and more and more of. So in this case we’re talking about shipments that may or may not need temperature protection. So these are things like mayonnaise, like beauty products, even LCD screens or car batteries. There’s a lot of things that at extremes, and it varies from product to product, extreme temperatures or in some cases maybe not so extreme, but the product is actually going to be damaged, damaged permanently, and may or may not have some kind of public safety risk.

So there’s two sides to this coin. One side is you’re using refrigeration or temperature protection when you don’t need to. In other words, it’s milder than expected and that’s essentially a waste of money. The other side of it is you’re not using refrigeration or temperature protection when you should be. And that risks losing the load, and that impacts service, and that impacts reputation.

So the solution to this is to be able to say for every individual shipment I do or I don’t need temperature protection. And the way that we do that is we’re simulating each route. It’s based on the location, the origin, the destination, time of departure, estimated time of arrival, the goods that are being shipped. And we’re modeling that route every 10 miles to produce a temperature forecast all through the route. And it’s taking into account not just the maximums, but it’s also taking into account the amount of time that the shipment is exposed to certain temperatures, because that’s an important factor, as well as of course the actual product and the temperature that product can or should not be exposed to. And ultimately all of that is boiled down into a risk score. And that allows a company to say, “Well, based on our risk tolerance, if the risk score is above this or below this, we will or we not use temperature protection or blankets.”

So to put that into practical terms, and this is a global CPG company, one that you’ll all know very well. And what’s interesting about this is their initial focus was really around the lost loads. Think of this like the proverbial iceberg. The thing that was obvious and that was seen were loads that are arriving that were obviously damaged or turned out to be damaged. And that was the problem that they wanted to solve. And of course in the back of their mind was transportation costs. But what ended up happening is we deployed these models for this particular company. And as I said, risk is opportunity. They weren’t really thinking about cost savings. But when we discuss this with them now, of course they’re talking about the fact that they’re no longer losing any loads, but the headline here is the amount of money that they are saving on transportation because they’re not using refrigeration when they don’t have to. And the more volatile that weather and temperature becomes, the greater the opportunity for these types of savings.

And the other thing that more recently they’ve been very focused on, as are many of our other clients, are reducing emissions. And reducing emissions is just, it’s an enormous challenge, but if there’s any low hanging fruit, companies really want to be grabbing that as soon as they can. And if you’re not using refri when you don’t have to, you’re reducing your emissions.

So as I said earlier, risk is also opportunity. And I’ll just call out one of a part of dashboards that we have. This is just something that we do to track cost savings in real time in the platform. It’s not cumulative. So this maps to essentially temperature volatility. Whenever it’s mild and unexpected, there’s opportunity to be saving money.

And really just to finish everything that I’m saying, this right here is always our aim. We always want to, rather than just talking about risk, or even measuring and scoring risk, we want to bring that back to the, “So what?” What decision is that driving? And what is the difference between that outcome versus what would’ve happened? In other words, what is the ROI from risk analytics?

Lauren McKinley:

Great. Thank you Jon and thank you David for the presentation today. And thank you to our presenters, David and Jon. Thank you to our attendees for joining us today. If you’re interested in learning any more about transportation optimization based on temperature, or any of the great weather insights from Jon and his team here at Everstream, please reach out to us at [email protected]. You’ll receive a copy of this recording sent to you within 24 hours. And thank you for your attending. We look forward to seeing you again in the future.