Keith Moore [00:00:00]:

When you talk about bringing AI to something like warehousing, there is a huge component of change management. People need to learn what the future needs to be sometimes and that's part of what we deliver as well. We consider ourselves partner to every organization we work with, not experts that need to tell people how to do their job.

Narrator [00:00:22]:

Welcome to Supply Chain Now the number one voice of supply chain. Join us as we share critical news, key insights, and real supply chain leadership. From across the globe, one conversation at a time.

Scott Luton [00:00:31]:

Hey, hey everybody. Good morning, good afternoon, good evening wherever you may be. Scott Luton and Jake Barr with you here on Supply Chain Now. Welcome to today's show. Hey, hey, Mr. John Wayne of global Supply Chain. Jake, how you doing?

Jake Barr [00:00:44]:

I'm doing wonderful. You know we're going to have a great time today.

Scott Luton [00:00:51]:

We are going to have an outstanding conversation here today, folks. We're going to dial it in on the logistics problem. No one talks about how disjointed tech is wreaking havoc on distribution. Now folks, I don't have to tell you our distribution centers and warehouses have no shortage of challenges. Think labor shortages, ever increasing customer demands and expectations, disparate automation and tech that doesn't play nice in the sandbox. All that and much more. But the good news, and there's always good news, if you go looking for it, there's a better path forward for your organization and especially our favorite part, your people. And that by and large is what we're going to be talking about here today with an innovative supply chain entrepreneur.

Scott Luton [00:01:33]:

It's doing great work out in industry, Jake. Should be a great and a timely show, huh?

Jake Barr [00:01:39]:

Absolutely. You know, it's wonderful when you get someone who is not only a technologist at the leading edge but a supply chain geek who really understands the work at a down on the floor level.

Scott Luton [00:01:53]:

I'm with you Jake. We have a very unique guest here today. And hey, if Jake Barr is singing your praises, you're doing something to write. So folks, stick around for a great conversation. It's going to offer up tons of actionable insights by the container load. So with all much ado, I'm going to welcome in our outstanding guest here today. Keith Moore is the chief Executive officer for AutoScheduler.AI where he is laser focused on bringing the future of technology into into the warehousing industry. Now in this role, Keith and a team work with a bevy of top 10 companies in the consumer goods industry, the beverage sector, distribution sectors and all along the way he's received numerous recognitions and accolades.

Scott Luton [00:02:35]:

Including being recognized, get this by Hart Energy Magazine, as an energy innovator of the year for 2020, man. Even better. He holds a variety of patents in the fields of neural architecture. I'm going to figure out what that is maybe today. And supply chain planning, one of our favorite topics. You're going to enjoy Lear learning from him here throughout the episode, I can promise you that. So please join me in welcoming in Keith Moore, CEO with AutoScheduler. Hey, Keith, how you doing?

Keith Moore [00:03:02]:

Doing great. Thanks so much for having me.

Scott Luton [00:03:05]:

You bet. So, Jake, we were talking pre show. I had a chance to connect with Keith up there at ProMat where they had 54,000 people. And I'm gonna tell you, Jake, you had to get a number and, and get valet parking to work your way in the booth at AutoScheduler. Have you seen Keith in action, Jake?

Jake Barr [00:03:22]:

I have indeed. I've used them for years and I swear by the product.

Scott Luton [00:03:27]:

And that's. That's. That's high praise, ring endorsement. I got to ask you, I got to say one more thing about Chicago, though, that we learned Keith was so busy that he didn't have a chance enjoy all the delicious food that the wonderful city of Chicago offers Keith. Is that right?

Keith Moore [00:03:43]:

So, yeah, what ended up happening is, you know, you wake up, ProMat starts early, you've got meetings in the morning. Right. Catch coffee with somebody, then you're at the show and you're bouncing, trying to connect with, you know, as many of the 54,000 people as you can. And then when the afternoon evening rolls around, you end up having event number one, event number two, event number three. And so I pretty much survived on hors d'oeuvres and water the whole. The whole show. Yeah.

Jake Barr [00:04:08]:

No way. No way. A techie in Chicago can always use door dash to get Chicago deep dish delivered right to your desk.

Scott Luton [00:04:17]:

Yeah.

Keith Moore [00:04:18]:

Have you been to McCormick Center? You get so lost in there.

Scott Luton [00:04:21]:

Yes, that is true. That is a major venue. Well, hey, if you happen to miss Keith at ProMat, take heart. You can still catch up with him and his incredible team at Gartner in May at one of the best supply chain trade shows around, Supply chain symposium. So stay tuned. We'll talk more about that towards the end of today's show. Okay, so, Jake, we're getting to know Keith. You already know Keith a little bit.

Scott Luton [00:04:44]:

I'm still getting to know him better and better. And I want to start with a fun warm up question we asked him earlier, conversationally, before we started the show. If he had six hours to himself on A weekend where he. He could. Didn't have to worry about business or family stuff. It was his time, he says. He said, number one, I never get six hours. But he said in fantasy land, if I did, he'd be smoking and grilling, hiking, and he wouldn't be taking any naps.

Scott Luton [00:05:12]:

He's not a nap guy. So, Keith, on that, in that, that hypothetical situation, what do you like to smoke and grill given the time?

Keith Moore [00:05:21]:

Man. So during COVID I actually ended up buying an offset smoker because I realized I don't actually need to smoke meat. I needed hobbies that weren't work or just spending time with my kids, which I love doing. Right. But I need hobbies for myself.

Scott Luton [00:05:35]:

That's right.

Keith Moore [00:05:36]:

And so with an offset, like a full brisket, I'm in Austin, Texas, so full brisket takes 18 to, you know, 20 hours if you do it well. So I, I do brisket burn ends. That's kind of something I like doing that you can do faster. You can do that in six to eight hours. And so that if. If I get to choose what I'm smoking, it'll be brisket, burnt ends, and the flat of the brisket.

Scott Luton [00:05:56]:

Oh, Jake, that sounds absolutely delicious. Huh?

Jake Barr [00:05:59]:

He's short changing me on meat, though. I'm just saying, you know, the burnt ends are great, but it's just a little bit of meat.

Scott Luton [00:06:07]:

You know, I just. We just had some burnt ends over the weekend at one of our favorite local places. And Keith and Jake, it is a thing of beauty, especially when you nail it. We'll have to get Keith's recipe later. One other thing, hiking. So your home base is Austin, Texas? Yep. If you were to get some time, get out hiking, where. Where do you like to go?

Keith Moore [00:06:29]:

So in Austin, because we're by hill country, there's a bunch of fantastic areas within 30 minutes. It's great if I have my, you know, full druthers on where I get to go. I'm flying to Calgary, I'm driving over to Banff national park, and I am taking a trip where all I do is hike every day.

Scott Luton [00:06:46]:

Okay, man. I like it. Jake, that's got your in ring of a. Your, your rubber stamp approval.

Jake Barr [00:06:53]:

I'll actually watch him vicariously, remotely while I st. Sit on the beach with a drink.

Scott Luton [00:07:00]:

No true words. Good stuff. Keith and Jake. It's important to find some kind of way to recharge the batteries and take a breath, especially at the pace of business these days. But let's switch gears. Keith, we got so much to get into you and your team been on the move in recent years. I'm looking forward to learning a lot more about your story. But let's start with some little extra context.

Scott Luton [00:07:22]:

Want to learn more about your professional journey, especially starting with what you did prior to the current venture. Tell us a little more. Yeah.

Keith Moore [00:07:29]:

So engineer by training, went to the esteemed University of Tennessee. So I am, you know, University of Tennessee, everything fan. Sorry, Jake. And so graduated, knew I wanted to start a company. Like that's been my career aspiration. I was the kid in high school that would like find a way to escape school, go buy a bunch of pizzas and sell them for a dollar a slice in the cafeteria. Like that was actually what I did in high school. So knew I wanted to go on that career path.

Keith Moore [00:07:55]:

Worked in tech for a few years, had a really cool job where I was doing this data acquisition control system work and realized that there was this big gap in the market where people had all of this data and nobody knew how to use it very well. And this was working with companies like SpaceX. Right. Cutting edge organizations. So in, you know, 2013, 2014, I got really into this new thing called machine learning. And it was I, you know, if you're a technologist, technically it is not new, but it was an emerging category thanks to compute. So ended up jumping on with a startup as one of the very first employees there here in Austin, Texas to learn how to start a company. Because I'd done a few side hustle companies where you'd started them.

Keith Moore [00:08:37]:

And hey, turns out I didn't understand product market fit or I didn't understand customer acquisition or I didn't understand the right way to do things. I realized, hey, I need to go learn from a, either a seasoned entrepreneur or somebody who was awesome at it. And, and that's exactly what I did. So I jumped on at a company called SparkCognition, went through Series A to Series D with them, became a unicorn. Focused on pure play, machine learning and artificial intelligence. So I've been in the space for over 10 years at this point, as you mentioned, it is a bit of an expertise of mine. After a while felt like I was ready to go start a company and while exploring a few different opportunities in Austin, ended up actually working with my family to start AutoScheduler.

Scott Luton [00:09:19]:

Well, Keith, we're going to get back to AutoScheduler in just a second, but I'm fascinated with what you shared there and Jake in particular, that gap in the market he was talking about, that sticks out and what he shared and then secondly, how he, he saw a gap in his own skill portfolio, I'll call it, and purposefully went out and not only created value for that company, but wanted to fill that gap for himself to go out and found and lead a company. But what'd you hear there, Jake?

Jake Barr [00:09:46]:

Well, what I heard was what I've learned about Keith for many years is it's one thing to have the brilliance of the technical insight. Right. The reality is he could translate that because he understood the mechanics of what the domain is in the supply chain space. Right. And he also appreciated, I think early on because of where he had a schooling, the disjointed silos of how the supply chain operated. Right. And so if you think about it, the new Greenfield frontier is the application of AI across all of these disparate silos that have grown up over 10, 15, some cases, 20 years. And quite frankly, one, they don't talk to each other, two, they don't have a way of being able to blend, bringing the information together.

Jake Barr [00:10:41]:

And three, they really struggle with trying to figure out what the consequence is of these flows. Right. And so that's where Keith and his team have really stepped into a void and are doing great things well.

Scott Luton [00:10:55]:

So that's a great segue, Jake. And I appreciate that, that void you are speaking of because I bet that, and that gap you, you identified out in the marketplace and, and your unique blend of skills and talents, I bet all that rolls up with some other stuff into why you started the business. Tell us more about that.

Keith Moore [00:11:14]:

Yeah, so for me, why the business got started is there was. And what we do is very focused. Right. There's a lot of great companies in supply chain. The supply chain tech world's fairly small. I've learned that since I've, I've come into it. And you know, we are really good at what we do because we are laser focused on a particular set of problems. And that, that problem is why we started the business.

Keith Moore [00:11:39]:

It actually started as a consulting engagement for Procter and Gamble. So the, the question was, you know, P&G said we have this problem at all of our facilities. We have these great performing sites and we have these poor performing sites. And this is across their plant warehouses as well as their distribution centers. Right. There's a distinction between the two. One has production, one doesn't. And the problem is, like, we don't know why we have spent all of the money on standardization of technology.

Keith Moore [00:12:06]:

So we have the same warehouse management systems running at all of them. We have the same process at all of Them how we run the sites is the same and we have the same hiring practices. So the people, the process and the technology should all be the same. But for some reason we have this big disparity in performance and some sites are unproductive, some sites are struggling with service levels. And so we came in and part of the realization there was the big unlock for them was in their management team. So the people, there's, you know, in a warehouse you might have hundreds of people. If it's a plant warehouse and you know, you might have thousands of people, there is a group of 10 to 20 people that make decisions on how they want to execute for the minute, the hour, the shift, the day and the week. And if those people are not in perfect sync on how they want to operate and how they deal with disruptions, you end up with a very disjointed operation that lacks productivity, struggles with service, et cetera.

Keith Moore [00:13:04]:

And so we started to solve that problem for P&G and can we just build a software that harmonizes all of these different data sources they have between, you know, warehouse data, labor data, manufacturing data, such raw materials and finished goods. And can we give them kind of the Lord of the Rings style one plan to rule them all on how they execute their workflow. So whether it's somebody who's 30 years on the job or 30 days on the job, that decision maker knows exactly what they need to do next and how they need to go about it.

Jake Barr [00:13:37]:

Now an incredible build on top of that, which again, because of his domain expertise, he understands it at a very granular, granular level is the so what, right? Because you get all that, you, you provide it to the individual, but you have to have context, right? And so the wonderful thing that I've seen with Keith and his team is they provide a context of here's what's going on, here's why the bottlenecks are occurring, here's how you could unplug that, that cog that's stopping the flow. Here are the trade off alternatives. And it gives them a holistic picture. Instead of someone trying to manage from their gut independently, location by location, site by site, even if they might have 7/10 of the information available to them at the time, they really need to make the call at one location and only 5, half of the information at the other one. So it, it creates a level playing field for how they're able to ingest and look at all of the changes and all the chaos that's happened.

Scott Luton [00:14:52]:

And Jake, doing it by the gut in your Example you shared isn't real scalable, is it?

Keith Moore [00:14:58]:

It's not.

Jake Barr [00:14:59]:

Now there's a great thing obviously. Look, we love the gut intuition, right? Because you build that up over time. The problem is it's variable, right? It's all subjective based on what that individual, he or she has had the opportunity to experience. So it's a great spot to apply artificial intelligence because you fill in the blind spots that they haven't had the opportunity to get whacked with. Up against the head to go, well, I'm about to make this call and if I do, I think it's a good one. But I'm basing it based on what I know, right, or what I can understand.

Scott Luton [00:15:40]:

So all told, Keith, go back and kind of finish put a period on the story about the why behind AutoScheduler.AI. It's if I've got the story right, you solved the problems that P&G were having across their ecosystem between distribution centers and warehouses and said, hey, we've got something here that we can productize and help so many other organizations that are, that are feeling these same pains. Is that right, Keith?

Keith Moore [00:16:06]:

That's exactly it is. We own the intellectual property for everything that we had built in the algorithms that we had built to solve these problems. And so it was, let's take this to market. If P&G has the problem, it's a guarantee just about everybody else has the problem, given how forward thinking P&G is in supply chain practice. And so we launched the business in August of 2020. So we are almost five years old. It sure feels a lot longer than that. I know it's hard to see on the camera, right.

Keith Moore [00:16:34]:

But the gray hair is real. But you know, so, so, you know, almost five years old as a business and we've been doing really well, growing primarily in that, you know, food and beverage manufacturing space, consumer goods manufacturing and moving into retail and grocery as well.

Scott Luton [00:16:52]:

It, it, it does seem, I can't remember when we first met, but it seems like you've been doing this like 25 years. So to think it, you've all you, you and your team have accomplished all this in five years. That is, that is remarkable. All right, so Keith and Jake, we got a lot more to get into. But where I want to go next is kind of the main central topic of today's show. Again, the logistics problem. No one talks about how disjointed tech is wrecking absolute havoc on distribution. So I want to start with this though, Keith.

Scott Luton [00:17:21]:

What do you point to in terms of why folks don't talk about this problem. Yeah.

Keith Moore [00:17:26]:

So disjointed tech, everybody, as soon as you say it, everybody's like, yep, I know what he's talking about. But when you look at the supply chain, you know, most companies think like, well, wait a second, I have an ERP, I have a planning system, I have a TMS, I've invested in, a WMS, I've invested in. You know, maybe I'm more advanced. I have LMS, YMS automation. You know, maybe I've invested in this real time transportation visibility thing, to use the, the Gartner term for it. Obviously, because I've made all those investments, it all talks together. And so nobody talks about it because everybody thinks that when they make an investment, it automatically plugs into all of the other investments that they've made. When the ugly reality is we still operate in a heavily siloed world where I could probably count, I won't say on one hand, but on two hands, the Fortune 100 organizations I've worked with that have a very good, centralized, harmonized data strategy on how to pull all of this data into one place.

Keith Moore [00:18:28]:

So it is kind of how everybody's gotten used to operating in the operation space. And people are so busy firefighting and logistics that they never come up for error. To think if we have all the data in one place from all of our systems, we wouldn't have this problem to begin with.

Scott Luton [00:18:43]:

So Jake, he paints that picture, that ugly reality. Your thoughts, Jake?

Jake Barr [00:18:49]:

The ugly reality is because of the size, scale and scope of what most supply chain organizations represent manage on a daily basis. They're so deep down in the woods of just trying to put out the latest fire and commotion. Building three is backed up four hours because of some logistical delay. Building two had a delay in the packaging materials arriving on time to support the finish operation over in building one. Right. Building one and three, the stuff has to merge in order to go out to do fulfillment. Wait, okay, well, what am I making my decisions off of? Am I making it off of independently the updates I got out of what's going on over and building three with the logistics delay? Am I doing it because I just gotten a production schedule update? Am I doing it because I just got, you know, oh, guess what, Susie, Joe and Jack are out on illness. And so we're eight trucks behind on loading.

Jake Barr [00:19:55]:

I mean, they're independent, but they're being managed today because of the chaos and the size and the scale as independent activities as opposed to realizing. No, no, they're a whole set of activities that you need to look at and make decisions over the top of or how to orchestrate the music.

Scott Luton [00:20:21]:

So, Keith, based on what you and Jake both are sharing there, this ugly reality, this disjointedness, this lack of harmony at the top of the list, what.

Keith Moore [00:20:31]:

Does it cost in organizations, the easiest answer is margin. Right? A supply chain. For most companies, some see it as a competitive advantage, but to all organizations, it is also a cost center. Right. And what we end up doing in supply chains to account for this disjointedness of tech is we end up building buffers in places we don't necessarily need one. The, you know, the optimal is like I say, I need a product at this location delivered right? Then there's a huge number of touches and flows that needs to be accounted for to get it there. And the intent is like, it just magically shows up. What ends up happening though, is we end up building these stages, these kind of safety nets inside of our supply chain to make sure that we're able to support that need at, you know, the cost of carrying inventory, at the cost of excess labor, at the cost of automation systems we never fully utilize.

Jake Barr [00:21:25]:

Cost of building out a building, adding additional doors, adding additional crewing and staff. All right, it's on and on.

Keith Moore [00:21:33]:

That's a great example. The number of times I've heard, oh, we have an overflow site over there, but we're going to shut it down next year. And then you call them in a year and you're like, hey, so do you ever end up shutting down that site? Well, no, we added another one. So I mean, that's exactly it is. You know, the supply chain pays for inefficiency in margin is what it comes down to.

Scott Luton [00:21:54]:

Jake, expound on that. The needless buffer, the, the needless safety nets, more staff, more square feet.

Jake Barr [00:22:02]:

To me, one of the Keith clearly pointed out, monetarily, it's no question it's margin points, but let's get real. The, the fundamental thing that we all deal with across the supply chain space today is the availability of talent to actually execute the work we need to get done done when it needs to get done right in an efficient way. And to make it a compelling piece of work that we don't have people like going, hey, I missed Johnny's baseball game this afternoon because I had to work three hours because all the stuff was screwed up and we couldn't figure out the right path to get it done so I could leave and go make the gain. So now I go home and I'm a heel. Because I'm missing an important thing which should have been on my off time. Right. So quality of work. There is a huge people element to it.

Jake Barr [00:23:00]:

It's not only training, it's the efficiency of the work and it's the work satisfaction so that you've got a team that doesn't hate coming in and punching the clock. Right.

Scott Luton [00:23:13]:

Yeah. Keith, Jake points on a great item there because that quality of work, I think he called it, where once you act on the opportunity that. That clearly you're. You're showing and illustrating the improvements that can be had. You improve that quality of work. And every day doesn't become an endless firefighting cycle where your people, your hardworking people that want to succeed feel obligated to wear that cape each and every day and miss family time. Keith, anything else you want to add to that? That's an important part of this, I think.

Jake Barr [00:23:48]:

Yeah.

Keith Moore [00:23:49]:

So I. I would add two comments and. And try. I'll try to ground it a little bit in reality as well, which is when you're in distribution and warehousing. So plant warehouse or distribution center. Right. Service can never suffer. That's always the objective.

Keith Moore [00:24:02]:

And that's. That's why this problem exists. To begin is like the objective of the supply chain is to deliver stuff where it needs to go, when it needs to be there. And these buffers and all these costs and all these challenges that have been built up due to the disjointed nature of the system, they wouldn't exist in a perfect world because everything would just get where it needs to go. And so you'd have optimized service at optimized cost. But in a warehouse. Right. That's really hard.

Keith Moore [00:24:28]:

And to make it really tractable, like a plant warehouse, for example, I might have a production system that's running and I need to get raw material to production, but I don't get real time updates on when the production schedule changes. And we're no longer in the world of production schedules changing on a weekly basis. They change minute by minute, potentially. So I need to know exactly what raw materials need to be brought to what line at what time. And I only have, we'll say, four pallet positions at the front of every line. So I have to do it in a really tight sequence. It has to be perfectly orchestrated to get everything there. And then if we don't get it there, production shuts down.

Keith Moore [00:25:04]:

Everybody yells at us. That's, you know, hundreds of thousands of dollars or millions of dollars that are lost. Right. That's just one practical example of data systems not being connected that ends up resulting in either a lack of service or significant addition of margin to the business or sorry, you know, overall cost to the business because we either add more space on the front end, we do a lot of things to avoid these problems that just better connectivity between systems would solve.

Scott Luton [00:25:32]:

And we're going to talk about a better way here in just a minute. We want to make sure we ground ourselves in reality. And really the next step in that, Keith and Jake, is I want to get you all to talk about the limits that traditional technology platforms and solutions. WMS, ERP, you name it, what limits they have. Keith?

Keith Moore [00:25:51]:

Yeah, so the best way to think of it is ERP and WMS. They are built and designed to track and catalog things. WMS particularly there's, you know, if you use the Gartner terminology, There's levels to WMS and some of the more advanced WMSs will also do directed work and, and task management to some degree. But they're not designed to optimize the overall constrained flows of a facility. Right. An ERP doesn't plan in size buckets less than a day, which to me still blows my mind.

Jake Barr [00:26:24]:

When you're building integer of 11 equals 24 hours.

Keith Moore [00:26:29]:

Right.

Jake Barr [00:26:29]:

A lot of crap that happens in a 24 hour bucket.

Scott Luton [00:26:33]:

Yeah.

Keith Moore [00:26:34]:

And like go to a warehouse and you're like, okay, this is what you're going to get done today. Well, what time do I need to get it done? Don't care. Right. And that's not how it works. And so you know, these systems have some, you know, ERP, that's a significant limitation, particularly when you get to the, the, I won't swear, but the get stuff done side of the business, which happens in distribution and warehousing and WMSs are really, they were initially designed for figuring out, you know, where do I have my inventory and what inventory is going on, what load and how do I make it easier to scan something and it tells me what to do. It's not saying that I have, you know, four different pick processes and they can only handle so much capacity at each time and they all merge at an individualized sort area. And so I have to make sure that I'm flowing things through the facility perfectly optimally so that no individual area gets overrun. It's just a limitation of the system because it's a bun.

Keith Moore [00:27:26]:

It's a bunch of advanced mathematics that has to think further than the next 5, 10, 20 minutes into the future.

Jake Barr [00:27:32]:

And play out scenarios Right, yep.

Scott Luton [00:27:36]:

So. And. And Jake, continue on with that thought, whether you're talking about.

Jake Barr [00:27:42]:

Yeah, fundamentally, I. I always view this as. I mean, in today's age, we're playing a giant game of neural network movement. Right. You're looking, I want you to envision, you're looking down into the box we call a warehouse or distribution center, and you're seeing all this myriad of movement and transactional stuff and flows and materials coming in and packaging and people movement. Right. And materials coming in and out of storage positions, going to docs and writing, accepting docs, and it's just mass chaos. And that's before I start dropping in changes.

Jake Barr [00:28:25]:

Unfortunately, the old adage happens, and it happens really frequently. It's as simple as a trucker gets delayed just one. Right. It's as simple as I run 10 minutes behind on one production line process, delivering materials that should be already at a palletizer and finished and they're not. Right. It's an unload coming in where, guess what? The guys were short one person today, and we didn't have it out so that it could be resequenced to be rerouted to the other side of the building properly. And we're. Unfortunately, we fall back into, as Keith was describing, the limitations of these individual elements that we're managing sole transactional tasks, as opposed to saying, I get it, I understand.

Jake Barr [00:29:24]:

But now I need you to take all these changes into account and tell me what's my best optimum to keep the place moving. Right. So that I can make the most out of the disrupted opportunity, as opposed to just handing out money, giving away.

Keith Moore [00:29:47]:

Margin points one call out, which is like, what is the best way to visualize the gap? So anybody who's listening, there's, you know, if you've been in a warehouse before, it's like we have the state of the art WMS and the YMS and automation and all these things, and then just ask the question, okay, if you have all of these things, why are the most used resources in a shipping office? Excel, paper, a whiteboard, and brainpower.

Jake Barr [00:30:10]:

Yeah.

Scott Luton [00:30:10]:

Right.

Keith Moore [00:30:11]:

And that's. So what's the gap? I think the gap is those four things that have to supplement, at least emphasized by those four things that end up being people having to supplement where their operational technology falls short.

Jake Barr [00:30:24]:

It's where the bowl of spaghetti attempts to come together with somebody putting their squishy hand in between all those moving parts.

Scott Luton [00:30:32]:

Oh, I love y'all paint quite the picture with analogy. That's right. And one other thing I heard There from, really from you both, is when stuff happens, because it does, right? Every single day. The snowball effect that as all these changes and dilemmas and delays add up, if we don't take a smarter, better, forward looking approach with the technology that's available, rather than using the very limiting traditional platforms that are out there, we, we only make it harder on our people. Right. That goes back to the earlier point, beyond the margin impact and, and you know, all the other data side of the tangible impact. Okay, so let's talk about Keith and Jake solo busting. And this may be you.

Scott Luton [00:31:23]:

We've been talking about this for years, which really goes to the heart of the opportunity we've got here. Little, little bit of a rhetorical question here, Keith, but how does visibility and that orchestration and that harmony that you're speaking to help eliminate those operational silos and create that powerful alignment that really can benefit the whole ecosystem?

Jake Barr [00:31:45]:

Yeah.

Keith Moore [00:31:46]:

So there's three stages and like I'd love to say these are Keith Moore's three stages of enlightenment in a warehouse. It's not, that's not the case. Right. This is a pretty standard process flow, which is, you know, visibility, predictivity, prescription. And you know, they go by a lot of different names across a lot of different types of analytics and AI. But the step one is like, you got to know what's going on. And you can't just know what's going on inside of your silo because you don't live in a siloed world as we just got done talking about. Everything that happens outside of your silo influences what happens inside of your silo.

Keith Moore [00:32:21]:

And so you have to have a single pane of glass that you can look at that tells you the current state of reality and goes as far into the future of like what you know about the future as it as possible. And in practice, what that means in a distribution or a plant warehouse environment is you need your production data, your warehouse data. So that's both, you know, finished goods and raw material, your labor data, your automation data, your location data, your trailer data, all in one place. So that at any point in a day you can look and you say, what, what is the current state of things? Right. This isn't getting into predictive, this isn't getting into prescriptive. But that, that's what you know. To answer the, the base question, I will get into the, you know, what is, what is prescriptive and orchestrated mean? I'm guessing in, in the future, but.

Jake Barr [00:33:12]:

My simplified translation of that is I need to Be able to do data ingestion from all of my siloed use units so that I can actually pull together the threads that hold them together and that will impact each other.

Keith Moore [00:33:28]:

One pane of glass. Best way to put it. Yeah.

Scott Luton [00:33:31]:

Okay. And that's step one, Keith? Correct?

Keith Moore [00:33:35]:

That's correct. That's just step one.

Scott Luton [00:33:37]:

Okay, do you want to. Let's go ahead and put a little more meat on the bone for all three steps and then we're going to dive a little deeper, especially as it relates to layering the technology on top or, or AI in a moment. But let's talk about step two.

Keith Moore [00:33:51]:

Step two is predictivity, right. If I know where we are, do I know what's going to happen? You know, if you're looking at a warehouse and you say, hey, I'm working 20 loads right now, all of them are due in the next two hours and I have 20 more that we haven't started working that are also due in the next two hours. You know, somebody looking at it could say, hey, we're probably going to be late on some of those. Right. So step two is being able to look at this data, look at known boundaries and conditions inside of your facility and start to predict this. Is that war gaming that Jake kind of talked about is you're doing the scenario modeling to say, given where we are and the objectives of the business, and every warehouse says, my objective is to maximize service and minimize cost. That's not true. Believe it or not, everybody's willing to make a service cost trade off at some point.

Keith Moore [00:34:39]:

It's really just how many dollars and what is the service level. And so you're doing that optimization to say what is going to happen and what are the most probabilistic outcomes Based on what I know now, the further you get into the future, the harder that is. Right. I can tell you what definitely what's going to happen in an hour. I can tell you without anything breaking what's going to happen in five hours or by the end of next shift. But if you ask me right now to tell you what's going to happen next Easter Sunday in a warehouse, not, not the one that's. Or you know what, however many days into the future, that is some 200, 300 something days, like, I have no idea. Right.

Keith Moore [00:35:16]:

And so like you need that second level of predictivity to identify what are the probabilities of what's going to happen in my facility.

Scott Luton [00:35:23]:

Keith, it's kind of like the hurricane modeling, right? Exactly. It's advanced, it's Incredible. What, what goes on now with weather prediction and especially with hurricane season. But to your point, kind of like what you're talking about, they can tell you where any given Hercules. Right. But as time, time goes further into the distance, that cone gets a lot bigger. Jake, we're talking about the first two steps here. Visibility and then predictability.

Scott Luton [00:35:49]:

What else would you add about step two?

Jake Barr [00:35:51]:

In step two, honestly, this is where you start to get beyond what I'll call even our best people and what they're capable of doing of a projecting. Because when you have so many bullets flying around you from all the changes, right. You're trying to absorb as much as you feasibly can and make a projection of the path to get the, the most work done, the most efficient way at the optimal outcome. Right. And so that's where that you're starting to augment that gut reality of well, I've had this happen to us in the past and what we've typically tried to work our way through is we do X, Y and Z, right? Well, maybe it's X, Y and Z to the power of two because these are the two nuances that you haven't been hit with before. Right. So this predictability piece is about being able to get out in front of the issues, right? Yes. I can go firefight all day long.

Jake Barr [00:37:02]:

And by the way, we have people who are masters at daily exceptional, one off firefighting. The problem is, strangely enough, you never get out of the firefight. It just is replaced by yet another burning bush. Right. So the objective here is to say how do I get us course corrected back towards what I'll call a norm? Should be. Right. And what's the path to be able to achieve that? How do I sequence all these pieces to make that possible? So that's that first piece.

Scott Luton [00:37:41]:

I like it, Jake. And kind of what you touched on there, one of many things my dear friend Kim Reuter was talking about. Yeah, we do a bad job empowering all the firefighters, but let's empower the fire preventers, right? Yep. All right, so Keith, we talked about the visibility, we've talked about predictability. What is the third step?

Keith Moore [00:38:02]:

And this is where AI kind of starts to come into play is that third step, you know, IBM definition prescriptivity, but I like to call it orchestration, which is that is where you are actually making decisions to optimize outcomes in the future.

Jake Barr [00:38:17]:

Future.

Keith Moore [00:38:18]:

So you're saying, and, and this is everything has a cost in reality. That's the case. Right. It's very often difficult to Quantify the cost. So if I go to somebody that's running a warehouse and say, all right, how late are you willing to make an outbound to get an additional pallet on that outbound? So if you're going to short a Walmart, are, are you willing to wait an hour? Are you willing to wait two hours? What about three? Eventually somebody's got a line, right? Nope, we'll eventually just ship it. And so this third step, this orchestration is now that we have all of this data, now that we're able to do this scenario modeling to figure out where the world's going to break and where my bottlenecks are going to be, I need to then start to make trade offs in decision to optimize outcomes. And at the end of the day, that optimized outcome is some combination of maximize service, minimize cost. Am I willing to ship something late to ensure that I have the labor to keep the production line running? And the answer is maybe, right, am I going to start loading trucks out even though they're not needed for two days because I don't have any space in the building? Those are the decisions that you end up having to make.

Keith Moore [00:39:23]:

And that's that orchestration layer on top, which is, you know, Jake mentioned people are generally pretty good at building rules on, you know, hey, don't ship things late. Make sure that we get all the product on there. But when you get all the way down into the weeds, right, you want to do that on an item by item level, a customer by customer level. You need to account for, for all of these. You know, this is a very complex multivariate time series problem to use some of the technical terms. And you want to be able to do that continuously because as soon as something breaks, what I think is going to happen in the future changes. Right. If a truck shows up late, I need to adapt my behavior to account for that.

Keith Moore [00:40:00]:

And that's what orchestration does. And that's really the AI enabled piece.

Jake Barr [00:40:04]:

What I want to be able to do out of this third piece and what they do so well is they create calm out of the chaos. I create calm out of the chaos because I have taken the prescriptive steps to create a cadence of activities where the people running the operations know what to execute next without having to stop and wait. Yep.

Scott Luton [00:40:34]:

Did you see there? And I love that calm out of the chaos. But did y'all hear there or folks listening or watching from home? Jake was providing his own rhythmic sound effects and I just heard the cadence of the. Of a Smooth organization there. Good stuff, Jake. So visibility, predictability, and then to use your word, which I like your word better. Orchestration, Keith. Love that. And we're going to be diving into more in future conversations on those three steps and a whole bunch more.

Scott Luton [00:41:04]:

All right? The golden aid, the darling poster child perhaps of this golden age of supply chain tech. And for good reason, right? Because beyond all the hype, it's making a massive impact. So, Keith, what is AI's role in all of this, especially in integrating, predicting, and optimizing distribution workflows?

Keith Moore [00:41:24]:

Yeah, so I have to define it first, which is AI by nature is replicating human behavior with computation. So the intent is we are replicating what a person could do at person quality or better. Right. With computation, which means it needs to be dynamic. There needs to ideally be some measure of change based on input. So data plus compute equals output in this case. And I realize that's vague, but when somebody says, like, do you use AI? They might mean generative AI, they might mean machine learning, they might mean advanced, you know, dynamic optimizations. And so the, the field of AI is really broad and as it applies to logistics for this particular problem of orchestration, this is what's called a combinatorial optimization problem, which you're like, okay, hold on, zoom out for a second.

Keith Moore [00:42:17]:

Which is, think of chess. So chess is a pretty simple game. There are 16 pieces you control, there's 64 squares on the board. By the time you are three moves into a game of chess, there are hundreds of millions of ways that game could have been played. And there are more ways to play a game of chess than there are atoms in the universe, which sounds pretty insane for the fact there's only 64 places you can put a piece and 16 pieces you control. Now expand that out to a warehouse and say, I have four thousands of pallet positions, maybe tens, maybe hundreds of thousands of pallet positions, hundreds of people that I'm controlling, lots of doors, maybe automation. We are playing, you know, the ultimate game of combinatorial optimization to say, what do I do next? Given there's an infinite number of ways to run a warehouse. And so what the AI is doing is it is actually looking through all of the different potential options for running that facility, not just for the next five minutes, but for the next day, two days, based on all known information.

Keith Moore [00:43:17]:

And it is coming up with, maybe it's not the absolute perfect plan, but this is going to be one of the 10 best options out of everything in the potential universe for running your facility. Oh, and by the way, when something breaks in five minutes, we're going to do it all over again.

Jake Barr [00:43:33]:

And I can't just do that by gut feel. Scott.

Scott Luton [00:43:39]:

So true. Especially my favorite part of that scenario Keith was painting with all the permutations, which I think is only the second time I've ever used that word I learned in math way back in the day is the dynamic aspect of that. Right. Because how cool would it be if we could set a plan on Monday and it still be great by Friday, you know, but that's, that's not reality kind of to what you are both were talking about earlier. So it's amazing. It truly is nothing short of amazing of how we're leveraging AI. Jake, comment though on the scenario Keith was talking about and where AI fits in and is powering this truly next generation. It's current generation but for a lot of organizations it's next generation stuff.

Scott Luton [00:44:24]:

Your thoughts Jake?

Jake Barr [00:44:25]:

I mean the truth is we haven't but simply scratch the surface of all the areas of how we'll employ AI to help us empower better operating results. Now the great news is what Keith has described is actually available now and it's an augmentation over the top of what I already have that's existing simplistically. Having run large scale supply chains, you know, the, the last thing I really want to do is spend a lot of time and effort kind of ripping and replacing a lot of the things that I'm using for those individual activities. Right. Because I spent a lot of money, time, resources getting it in place. Right. Teaching people work processes also. So if I'm able to use the power of AI in this case to actually augment those processes where I don't have to go back and rip and replace, but I get the benefit of using the power of it to improve my outcome and my, my results.

Jake Barr [00:45:36]:

That's heaven.

Scott Luton [00:45:38]:

Yes, undoubtedly it's getting our cake and eating it too. And Keith, to steal from Jake, there it is available now we're not talking 2030, we're talking right there. So I want to bring it all together right. Because we I'm going to ask you a couple of questions that we've been kind of speaking to a little bit piece by piece throughout the conversation thus far. But this is what you and the AutoScheduler team do?

Keith Moore [00:46:00]:

Yep.

Scott Luton [00:46:01]:

Right. So what does a real intelligent dynamic logistics platform look like in action and tell us more about not just the tangible benefits but the intangible benefits. Yeah.

Keith Moore [00:46:12]:

So what, what does this look like in action? Right. This Is a layer on top of existing systems is the first call out, right? The whole intent and how we started this was how do we harmonize all of the data that we already have. Most people don't want to implement new systems because it's like open heart surgery on their existing operation. That's not the intent whatsoever. So you're taking and harmonizing all of this data across all of your systems. If you were to deploy something like an AutoScheduler on your network, it is then continuously dynamically running. It understands exactly how each site needs to operate, right? Site number one may stage, site number 2 may not. Site number three has a weird production setup and site number four is in a rectangle.

Keith Moore [00:46:51]:

And so we need to account for all of this uniqueness of doors and the fact that some are, you know, automated loaders and some aren't. Right? It's configurable to where when we're modeling a site inside of our platform, it is tuned to that site to understand how that site operates and how it runs. It is then continuously and, and you know, continuous in nature is, you know, some sites every 10 minutes, some sites every hour. But it is looking at all data available, building that plan, identifying where the bottlenecks are, telling you exact, you know, the what, when, where and who. So what do you need to do? When do you need to do it, where does it need to be done and who is going to be doing it, whether a person or automation. And then the optimal is this is fully orchestrated. So we have some sites that even have this, you know, harmonized data advisory control layer that people can make decisions off of. And some of them are even just automating straight back into existing systems.

Keith Moore [00:47:46]:

We'll automate yard moves, we'll automate waving automation, work queue management inside of the warehouse management system. You go from we used to hand out work inside of a facility to you know, we're moving towards the warehouse of the future which is, you know, the man and the dog. The man's there to feed the dog, the dog's there to make sure the man doesn't touch anything. And so people inside of facilities are able to spend their valuable time on fighting actual fires and not on the management of overall orchestration of work. And that's in, you know what, Everything I just said outside of the man and the dog is insights today. So, so this is deployed, this is a done deal more network wide at you know, companies like P&G, companies like PepsiCo or Frito Lay specifically now. So we have a Lot of really good use cases and proven outcomes at this point.

Scott Luton [00:48:34]:

So what you're saying there, Keith, is you'd love to talk with the folks that want to kick the tires and be good skeptics, constructive skeptics, because you've got the proof.

Keith Moore [00:48:44]:

We do.

Scott Luton [00:48:45]:

And I bet you welcome those conversations, don't you, Keith?

Keith Moore [00:48:48]:

Certainly do. We have a great team and we love what we do, right? We're really good at it. My team is extremely good at not just the delivery of the software, but the change management tied to it. Realizing when you talk about bringing AI to something like warehousing, there is a huge component of change management. People need to learn what the future needs to be sometimes, and that's part of what we deliver as well. So we consider ourselves partner to every organization we work with, not experts that need to tell people how to do their job.

Jake Barr [00:49:20]:

Scott, I've been at it for multiple decades and I can tell you one of the other things is it, it's a perfect illustration to me of, you know, a principle I live by about saying, hey, I've got to be on a constant learning journey, right? Fundamentally I learned by the use of the, the capability how to look at how I set targets and objectives between operations differently. Now you think about it, you step back and you go, wait a minute, I got 30 plus years of experience, I've done, I mean I can do operate all the equipment, I, I got the gut instinct I built from all the possible failures and all that stuff. And yet I'll tell you, I was like a kid in kindergarten when I'm actually learning for the first time, what were the nuances that were the reasons, the real reasons why performance in first location versus fourth location were distinctly different, even though they had same throughput, same size facilities, same product makeup, same customer mix, all that stuff and you're going, I should have known that, right? I mean I, I should, it's just too much, right, to be able to absorb and try and delineate, to get it down to that, those kind of tangible realities.

Scott Luton [00:50:45]:

It takes a different approach, a different technology, a different way of looking at your ecosystem defined Jake, as you were saying earlier, that calm out of the myriad of chaos and it's so, so true. There's a better way. Keith, I really have enjoyed the picture you've painted here today and, and again, we're going to, we're going to make sure folks know how to connect with you and the AutoScheduler team in just a second. But before we do that and before we get Jake's patented key takeaway. I got to ask you as a fellow entrepreneur know we got a strong entrepreneurial contingent within our global audience, which is the smartest in all the world or, or if they're not active entrepreneurs, they, that's what really what they aspire to be. Much like you. Earlier in your journey, if you had a piece or two of advice that you, you would extend to those folks that are in the trenches or they're, they're trying to come up with the right idea or whatever it is, wherever they are in their entrepreneurial journey. What would that advice be?

Keith Moore [00:51:42]:

Yeah, it's, it's foundationally simple, but it's something that somebody told me and it's part of why AutoScheduler has been successful so far, which is find a problem someone is willing to pay you to solve and solve it. A lot of companies are like we're going to go build up this massive technology and invest very heavily and raise all this money. And the reality is like, particularly in logistics and supply chain, people pay for solved problems full stop. And so starting a company can be a consulting engagement to say, what are your problems? Here's my expertise. How can I build you something that I can retain the intellectual property for that will solve your problem? And you use that as a springboard, you can iterate on that. It doesn't have to be this massive capital intensive thing that you end up doing. And particularly given the markets these days. Right.

Keith Moore [00:52:34]:

Raising capital is extremely difficult right now. And so really would advocate for anybody trying to start a company, starting small, finding somebody that's willing to pay for you to solve a problem. Go do that. Figure out how to rinse, wash and repeat.

Scott Luton [00:52:49]:

I love it. That sounds like billion dollar advice to me. Find that problem first. It's not. And I love this beautiful simplicity in that it doesn't have to be the chicken and the egg when it comes to problem and solution. Find the problem first. Jake, before we make sure folks want to connect with Keith, you want to comment on that? Billion dollar advice from Keith Moore?

Jake Barr [00:53:09]:

Spot on. I mean fundamentally supply chain is gainful lifetime employment because there's always another problem. As soon as you fix one, there's another one. And it's because we, we live in a dynamic business environment. Markets change client channels through which we sell, change product mix, product innovation, all those things influence. So it's a, it's a concept mixing bowl, right, that you're trying to put together and figure out what to do with. You know, my, my fundamental takeaway for today is just hey, it's, it's refreshing to actually see the ability to simplistically apply new age tech in a way that makes operational sense.

Scott Luton [00:53:58]:

That sounds like an excellent, terrific key takeaway here today. A patented key takeaway from Jake Barr, the John Wayne of Global Supply Chain. Well said, Jake. Keith. And you know you are as advertised. When we were getting in here earlier today, Jake said, I tell you, Keith, Keith is found that at that unique intersection of folks that get the brilliant technology can talk the technology talk, and better yet, they know how supply chain works and that that intersection can be found some amazing.

Jake Barr [00:54:30]:

He's a supply chain geek. He really is. That's. I mean, it's the amazing thing when you can find people like that who actually know how the work is done in a, in an operation.

Scott Luton [00:54:43]:

Yep.

Jake Barr [00:54:44]:

You're no longer a techie, you're part of the fraternity.

Scott Luton [00:54:47]:

Oh, I love that. And Keith, you, I bet you embrace that supply chain geek term of endearment, don't you?

Keith Moore [00:54:53]:

I do. It's actually a requirement at our company, which is everybody has to know how the business works.

Scott Luton [00:55:00]:

Oh, I love that.

Keith Moore [00:55:01]:

Whether you know how the tech works or not, you need to know how the business works. So we actually, we have constant sessions with, you know, engineers that work on back end technologies on how to differentiate when case and layer pick setups need to be changed.

Jake Barr [00:55:15]:

Right.

Keith Moore [00:55:16]:

So yes, that, that is innate to who we are as an organization.

Scott Luton [00:55:19]:

Okay. I love that. That kind of reminds me of Yossi Sheffi. He was writing here recently about how no matter what business you're in, every manager, every leader would be best for him or her to know how supply chain operates. And I, I tend to agree with him. Okay, so Keith Moore, first off, how can folks reach out and connect with you? Let's talk about Gartner in just a second. But let's get the basics first. How can folks connect with you, Keith?

Keith Moore [00:55:45]:

Sure, add me on LinkedIn. Easiest way. Keith Moore, CEO of AutoScheduler.

Scott Luton [00:55:50]:

It's just that easy. Okay. And you are going to be at Gartner Supply Chain Symposium along with me and Jake and the whole gang. I bet the whole crew. AutoScheduler.AI crew is going to be down there. That is, let's see, that's May 5th, 6th and 7th in Orlando.

Jake Barr [00:56:06]:

Yep.

Scott Luton [00:56:07]:

And to all of our wonderful audience members out there, if you're there, if you're one of the fortunate 4000 or so lucky souls there, Keith, I bet you'd love to share a cup of coffee or an adult beverage, huh?

Keith Moore [00:56:19]:

I would so if you're going to be there, you know, please reach out to me. I'd be happy to spend some time, whether it's just talking about your problems or obviously would love to tell anybody.

Jake Barr [00:56:27]:

About AutoScheduler or in Barcelona.

Keith Moore [00:56:31]:

Or in Barcelona. That's correct. Later in May.

Scott Luton [00:56:34]:

Okay. You're going to be at both, Keith, is that right?

Keith Moore [00:56:36]:

Yeah, it's. It's going to be a busy spring for me.

Scott Luton [00:56:41]:

All right, send pictures. I'm gonna see you in Orlando. But I have not been had the good fortune to make it to Barcelona yet. So I'm jealous of you and Jake. But I look forward to hearing all about it. Okay, so big thanks, Keith Moore, CEO with AutoScheduler.AI. Folks, reach out to Keith on LinkedIn and Keith, your URL. Where can folks find you on the web?

Jake Barr [00:57:04]:

AutoScheduler.AI.

Keith Moore [00:57:06]:

Oh, very easy.

Scott Luton [00:57:07]:

Yeah, it's just that easy, man. Well, Keith, always a pleasure. It's great to reconnect with you here after a great session in ProMat. And Jake Barr, really, really enjoyed your perspective on Keith and what they're doing over at AutoScheduler. I really enjoyed you being here as well today.

Jake Barr [00:57:23]:

Anybody that can actually take my gut instinct and actually make it better is even better.

Scott Luton [00:57:28]:

Oh, man, the praise is effusive here today. Keith Moore. Well, hey, I look forward to seeing y'all both at Gartner in Orlando and to our audience. Hey, I hope you enjoyed quite the conversation over the last hour. There are tons of actual insights and better yet, we answered why from about a million different perspectives here today. Why we've got to change how we use and leverage and deploy tech out in our warehousing industry. Distribution centers, fulfillment centers, you name it. Reach out to Keith.

Scott Luton [00:58:00]:

You, you will not, you won't be disappointed there. But most importantly, your homework is this. Take one thing you heard here from Keith or Jake. Put it into practice. It's all about deeds, not words. Let's change. Our business is done, if for no other reason than for our people out there. But with all that said, on behalf that Supply Chain Now Team Scott Luton challenging you.

Scott Luton [00:58:19]:

Do good, get forward, be the change that's needed. And we'll see you next time right back here on Supply Chain Now. Thanks everybody. Join the Supply Chain Now community. For more supply chain perspectives, news and innovation, check out supplychainnow.com subscribe to Supply Chain Now on YouTube and follow and.

Keith Moore [00:58:38]:

Listen to Supply Chain Now. Wherever you get your podcasts.