Best Co-Worker Ever: Partnering People with Technology for the Warehouse of the Future

In this episode, we sit down with Carhartt to discuss their rapid growth and how challenges, such as capacity constraints and keeping up with heightened demand, were addressed with an assisted-picking robotic solution and a partnership with DHL Supply Chain. This episode was recorded live on the show floor of a major supply chain industry event.

Special Guests:

• Tony Gariety, Vice President of Logistics and Distribution, Carhartt
• Omer Rashid, Vice President of Operations Development, DHL Supply Chain

Transcript

Copy Transcript Copied

Will Heywood 0:09

Welcome to all business, no boundaries, a collection of supply chain stories by DHL supply chain, the north American leader in contract logistics. I'm your host Will Heywood. This is a place for indepth discussions on the supply chain challenges keeping you up at night. We're breaking beyond the boundaries that are limiting your supply chain. This episode is best coworker ever partnering people with technology for the warehouse of the future. We are joined today by Tony Garty, vice president of distribution and logistics at Carhartt and OER Rashid vice president of operations development at DHL supply chain. Let's dive in. So Tony, why don't we start with you?

Tony Gariety 0:50

Yeah, I'm Tony Gariety , um, vice president of the par distribution operations been with Carhart about six years now, prior to that started my career at the LT industry. And most recently spent 20 plus years in the supply chain sector.

Will Heywood 1:05

Great

Tony Gariety 1:06

Home improvement sector, I should say. Mm-hmm <affirmative>

Will Heywood 1:09

Big box store .

Tony Gariety 1:10

Big box.

Will Heywood 1:11

Yep . Yeah . Good Omer.

Omer Rashid 1:13

Yeah. Thanks. Well , Omer Rashid , vice president of the solutions design with DHL supply chain been with DHL for over 11 years now. And prior to that, another 10 years in logistics consulting and we designed solutions for warehousing transportation, packaging solutions for all of our customers, including heavily automated sites, robotics, digitalization solutions that can bring heavy improvements for our customers operations.

Will Heywood 1:38

Good. Well, welcome to you both. And thanks for being here. So yesterday you guys gave a really good breakout session, big crowd, all of that, but one of the things that I was sort of taken by Tony was the history of Carhart. I'm familiar with the brand as a consumer, but I didn't know all of the history and kind of how you've gotten to where you are today. So would you mind just kind of stepping back through that for us? Sure.

Tony Gariety 1:59

Carhartt we're a , a privately owned single family, fourth generation company that started out in , uh, Dearborn, Michigan by Hamilton Carhartt. And he had a , had a simple vision of how he got started. He actually was just observing railroad workers, wearing overalls that were essentially rags that were produced by the prison system back in, this is 1889. So we've been around for a long time. And his philosophy on that was, you know, these workers deserve more and I can do it better. Mm-hmm <affirmative> and that's been, you know , a commitment and a mindset that has been embedded in our culture. It's still here today. And so, like I said, we've got a long, Heritage's been around a long time, really experienced a lot of major growth in the eighties and nineties, just when we, the work wear started to grow out of the traditional overall setting that we started in and then through social media and, and those awareness, you know, issues, and popularities, we've just continued to grow and really are seeing some, some very large and sustained growth. And , you know, those are obviously some good challenges to have, but certainly challenges we're looking to meet the meet the moment. Okay,

Will Heywood 3:08

Great. H how are we here together? So you had some challenges you said, and I think that's when the DHL relationship happened, walk us through what was front of mine for you you needed to kind solve for and how you started the partnership up.

Speaker 3 3:23

Yeah, so kind of rewind back to pre COVID . We were operating two distribution centers, one in Hanson , Kentucky, that's called the R CV , which is our automated flagship distribution center. It's automated our own. Like I said , our own employees. We have operated that DC since 90, I think it's 96. And then we had a three PL in , uh, in Memphis, you know, those were, we were kind of keeping pace somewhat, but our growth over the next few years. And then when we went out and looked at our demand, we still had substantial growth, but there was a lot of unfulfilled demand that we just weren't beating . Mm-hmm <affirmative> . And so our challenge was how do we close that demand gap, get out ahead of the growth curve and, and be able to, to , like I said, really meet the moment. And so we had to go out and my challenge to solve was, you know, we really needed something that helped us and speed the market. Mm-hmm <affirmative> , you know , how can we get there? And , um, so we went out and looked and went through a lot of partner evaluations and thing where the partnership with DHL was, was pretty obvious with us was, you know, just the , the size and scale DHL, the business model, the , the business expertise in this space. And we're really impressed with the technology and visualization, not only things that you were doing, but the effort and commitment around that, mm-hmm , <affirmative> really led us to believe that DHL's on the cutting edge of these things to just stay out ahead of it and would really help us in that speed to market and close that close that gap. So, so we went into the partnership and we opened a , a facility back in , uh, 2021, and we're slated open two more this summer. And then we've got a , a big automated center slated for next summer. So partnership to , you know, is , has been fantastic. And then when we, we actually opened that first facility manually, just because of the, the supply chain timeline, you know, everybody's experiencing, started getting elongated. So he kind of had a choice of, did we wait a little bit for automation? We'd go ahead and let's get started. And so , um, so we decided to go manual, but again, one of the great problems was we had a lot more demand that we had initially planned for that created a little bit of a backlog, basically starting up the operation. But our , um, our , our DHL partners come to us and says , Hey, I , we have a technology. I think we can get in quick, that'll help us increase the order, assembly capacity, improve some productivity, and really help us work up to that volume that, that we were seeing so soon. So , you know, that's when we, we went in with the locus , uh, the locus solution and, and it really has paid, paid big dividends. I mean, you know , to this point, we've, we've seen as much as 80, 90% improvement on order assembly and, and over 50% just straight picking productivity. So really helped us close that gap and have one the biggest years we've ever had, which falls our peak season. So that was, that was a season I'm referring to. Yeah . So really helped us close that gap, meet that moment.

Speaker 2 6:28

Okay, good. I wanna come back to the locus and automation topic in a minute, but maybe Omer to you. You know, I think we're lucky where we've worked with Carhartt type companies that have these challenges where they do experience this big growth. Can you talk a little bit about how DHL has prepared itself to, you know , kind of answer the challenge that they're facing and help them out through these big demand spikes?

Speaker 4 6:51

Yeah, I , I think there's a lot of different things involved with that, right? With a lot of folks that come together and over the years, as we've been working specifically in e-commerce operations or where we have different seasonal challenges, we worked on labor programs that help us bring in folks, train them, get 'em up to speed, so that they're ready to deliver for those seasonal spikes. And then as we grow with the business, they can stay on and help support that growth. I think when the challenges we've had, especially in 2021, Tony mentioned is, you know, labor was, was really challeng for us during that the height of the pandemic. And so what we had done previously with regards to robotics and our solution with locus, making sure we had integrated, tested it out , uh, understood the pros and cons, and also what it took to stand up a facility quickly with , with a locus type solution, helped prepare us for the labor shortages that would come as well as the growth that we would need to handle. So, you know, being ready to, let's say, plug in robotic solution into manual operation really prepared us for this type of situation and helped us get our heart to where they needed to be, not just from a capacity standpoint, but , uh, service level expectations during those peak seasons as well. I think the robotics help us with cycle time when we're fulfilling orders, especially when you have priority orders. And , and so just, it's a lot of the work done ahead of time with our organization to work with our solution provider partners of technology partners, so that we are ready to go when a CU comes to us with this type of challenge. Yeah. And I think just if I could build off of that, you know, we're doing it ahead of these solutions necessarily be ready to be productized. So we're working with those partners to make sure they fit in our business and then we've identified, you know, our customers problems. So that one we're building up that solution. We know it addresses not just one customer, but multiple customer solutions and we're ready to deploy rapidly when we need to,

Speaker 2 8:49

You know, we're on a trade show floor and there are a bunch of different vendors out there. And if you go to like a proma, there's even more, and there's just sort of an avalanche of providers coming into this space, how do you know what to recommend to a particular customer?

Speaker 4 9:03

Yeah. I mean, it's a great question. And it's the magic question. That's one we are challenged with consistently. I think what we're really good at doing, or what we have experienced at doing is understanding where a particular use case in our business is first and foremost, right? Where do we have the biggest challenges and what do we need to solve for then when we're engaging with technology partners, because there's a lot of them out there, as you mentioned, you know, we can identify those that are looking to address those problems or on the cusp of addressing them and help them get there with that. So we have an innovation funnel that we use, where we take ideas, innovation, ideas, and we're able to identify where to match those with vendors. We might see on a trade show floor and then develop the solution with the vendors and make sure that it applies to the use case for two. So we're doing the research up front . I think there's a value there to that so that our customers, you know, they may have 1, 2, 3 distribution centers. We have almost 500 in us and Canada. So it's imperative on us to understand where those solutions fit so that we can get the most benefit from them. And then us being able to do that again helps with our customers not having to go do that research and sift through all those vendors. And I think just our experience comes into play there too, as to knowing who a good partner is and knowing what's successful or what will be successful for their business.

Speaker 2 10:24

Yeah. Yeah. Tony, you mentioned you started up manually. Did you know much about locus or what was your experience, why did you decide to automate, did you have doubts about these types of solutions? How did you kind of get to where you are now?

Speaker 3 10:38

A little bit of familiarity with certain applications of it? So had had a little bit of experience with it. AGV technology, you know, nowadays is not even, it's not even new anymore. Almost it's becoming pretty, pretty commonplace, but, you know, just trusting the, the DHL team touring some of the sites that already add it , you know, see it in , in action . And the fact that, you know, one of our, our partnership strategic moves was to integrate to DHL's warehouse control systems and all of that, and actually leverage that size and scale. So no one that it's already been integrated it's in, it's proven they're the ones using it and tried and true and certainly fit what I felt like fit our application. The product may not be straight up, you know , super small econ type things, but it fits up and down our profile. So, so it's kind of a combination of their experience, what I saw with where they were using it, how they were using it , uh, against what, what we needed. And then, and then two, just kind of playing back to that old warehousing 1 0 1, when you see these big footprints, a lot of skews, a lot of people, a lot of walking, you know, it just, it just really make made good, common sense. And how do you , how do you cut down the travel, right? And that's one of the largest labor hours of, of any distribution, warehousing operations. So it's kind of, kind of one of those age old challenges that every big warehouse has. So, you know, really, really see that.

Speaker 4 12:06

And if I could build off that sure will . And Tony , I mean, great point. I think it's important to note that we're not just plugging in a solution because we like, and we've used it elsewhere. There's a lot of emerging technologies that are trying to solve a similar problem. And so there may be one or two that fit a certain profile and then others that a different one trying to solve the same problem, right. For the each making problem. I think what's important is working with Carhart. You know, we did have to look at that profile information, look at different options that fit , uh, for that type of solution and then land on the locus solution. Mm-hmm <affirmative> whereas we were prepared to deploy it then because we were very familiar with it. So there is still some process of making sure we've got the right picking solution in my phone , but then we're ready to go quickly when , when we've identified it.

Speaker 2 12:52

Yeah. So this term profile, we use a lot in our day to day , but for the audience, if you could kind of break down what are the variables that sort of make up a profile, it would make one different from another. Yeah.

Speaker 4 13:03

So, you know, we get pretty detailed. So we look at, you know, order history data, we try and look at projections of where the orders are gonna come, where they're gonna go to what they're gonna look like. So when we talk about profile, talking about order size is the order size, is someone ordering one or two units? Mm-hmm <affirmative> when they order online or we shipping to a wholesale customer that might have 50 units. And then how far are those units spread across our skew base ? Right? So could it be coming from 20,000 SKUs? So I have to go pick 50 lines from those 20,000 skews for those orders, or am I picking five skews every time I go to a line to pick for a wholesale order. So it's really, the profile would be number of lines per order, the number of units per line. And that's really what we look at. And then based on that, how big would our traditional pick line be? And then we can start to examine or engineer a better solution that gets us reduced pic line length, reduced touches and things of that nature. Okay,

Speaker 3 14:01

Good. And even to add to that is , you know, product, weight, dimensions, correct characteristics of your product, you know, all those really factor to , you know, those profiles and like home Omar was alluding to, you know, we've got a lot of different profiles. You know , we have retail stores, we have, you know, small retailers, we have huge retailers, we have a E eCom space. So we got a little, little bit of the horizon, but like , to your point, it's really about matching up customers needs the profile of their yeah . The essence of the work they're doing and do the two .

Speaker 2 14:36

And you're fulfilling all those channels out of the same warehouse infrastructure. So you don't have a dedicated building for e-commerce versus wholesale.

Speaker 3 14:43

We don't today. I mean, as we continue to build out our, our network and size, there's gonna be some opportunities for service to probably do a little more isolation. Mm-hmm <affirmative> geographically for the, for the.com space. But primarily we're, we're looking to take a lot of our high moving items and replicate those across the network and then centralize some of our slower moving the more, the long tail .

Speaker 2 15:07

Yeah. Got it.

Speaker 3 15:08

Things. But today we don't have a specialized warehouse per channel or flow or an account it's serving basically all

Speaker 2 15:17

Customized . Yeah. Right, right. So you talked about the travel, I'm always kind of astounded when I go into any of our warehouses, the size of them. Can you talk about how big are they and as a general associate, what would my day look like if I did not have an automated

Speaker 3 15:30

Yeah. That's , uh, that's a good one. Um , most of the warehouses we're operating today are between anywhere from a half a million to 850,000 square feet today. Um , the one we're looking at for next year's and it's actually 1.2 million mm-hmm <affirmative>, but yet the associates on an average day, I mean, literally could walk 10, 12 miles in a , in a warehouse that's not unheard of mm-hmm <affirmative> depending on obviously how you've got it laid out. And that , that produces challenges for us to, you know, figure out how to best lay out the warehouse to minimize that. But there's still a, a physical nature of having to, to walk a certain amount. So, you know, how do you, how do you take a big space and shrink it down fundamentally? And this is where the solution will help us. Also, as we get into some, some pick and pass where an associate may be doing a they're still walking, but they're walking in a short, short space. So it's actually cutting down travel time and they're doing more picking than traditionally pushing a cart or walking. And if you would ask an associate of the three physicalities or the physical nature of a , of a picking processing job, it's the picking, the walking and the pushing. Yeah . And they they'd all tell you the walking and the pushing are the , the two things that take its toll. So, and we would certainly rather have folks picking than walking and pushing. So it's that proverbial sounds cliche, but it's truly the win-win, you know, you're taking the physicality out of the job. Honestly, there's a lot of safety features in that too with cause our , our carts that we have at some of our other centers, I mean, they're , they're 75 pounds just far. Yeah. So, you know, that's that's um , and then back to just straight up labor play on travels, historically, one of your, one of your largest labor hours just in the

Speaker 2 17:30

Walking . Right. So Omer , um, we've talked a bit about locus on other podcasts, but just, can you describe what it's like as an associate? You know, what are they doing? What are the robots doing? What are the kind of primary differences between activity and an automated warehouse versus one that's manual?

Speaker 4 17:49

Yeah. So as Tony described in a manual warehouse, traditionally, we would set up a pick cart bigger than a shopping cart, but one where you could fit multiple totes, you know , try and pick multiple orders, reduce the travel that Tony talked about. So a large cart say would tend to 12 totes on it. And in a manual operation that picker would snake their way through an entire pick line, pushing that card , stopping at each location, scanning each location, each item , identifying which tote to put the item they picked into. And so that can be quite a laborious process, as we said, and with locus, it's actually changes the paradigm a little bit because it's not just automating the travel locus actually provides a solution where we have multiple robots per pick. So our ratios around in order from three to four robots per picker, and what the robots do is they're directed two pick slots within zones within the warehouse. And they may have multiple order totes on the robot already inducted into them . So they'll navigate their way to the pick slot. Uh , within a zone, you'll have a picker that could identify a robot. They want to pick to , uh, via Bluetooth. And the robot on an iPad will display the item, the location and the quantity that they want that selector to pick. So now you've got a visual cue . Uh , the robot has stopped at the location, so there's not as much searching for the associate. And it's as simple as picking the item. We can scan it using the iPad, that's affixed to the locus, and then they're depositing it and confirming on the iPad where they've picked that item. As that robot departs, the associate can turn around and another robot will present it itself from the same zone could be for a completely different order. But now that that picker is not walking and the robot is the queue as to where their next pick slot is. So, you know, if you combine many pickers in a forward pick area, they're working with many robots. So again, it's become very efficient and it's what locus calls a swarming technique to, to select the orders. What's really nice about it is if you have a rush and you need to go get one order out, you can send one robot out to get that work done. And it doesn't impede the rest of the order way that's being picked. So you don't have to wait for one order to be picked for the next one to be in. And that's really part of the beauty of the solution as well.

Speaker 3 20:07

Okay ,

Speaker 2 20:08

Good. You , you talked a lot about employer of choice yesterday and with the labor market, what are all the dynamics and how the locus solution helps support what you guys are driving

Speaker 3 20:18

At? Yeah, I mean today , um, you know, as anybody of the , especially the past few years of labor market has just become, you know, exponentially competitive. And to me, you know , we wanna be the employer of choice. We want people to come to work there. And so how do you change the whole traditional environment of the warehouse mm-hmm <affirmative>? So I spoke to earlier, you know, the physicality of the job of the walking and the pushing, if you can take that outta the equation, not completely, but significantly the nature of the work is a lot more appealing. Mm-hmm <affirmative> not , not straight up its manual, but two, when you start thinking about the training, I mean, it literally would take you weeks to get somebody used to the locator system within the warehouse. And then you're teaching 'em variety of archaic scanners and other things, and, and being able to bring in a technology that literally just makes sense to people. It's what we talk about, the , the interfaces, an iPad, well , everybody last 20 years has grown up with an iPhone in their hands. So it's just, it's natural. And so with the queuing that Omer had discussed, and it really takes a lot of the unknown part out of the training and people literally can go to work in , in the matter of minutes and hours mm-hmm <affirmative>. And so that's a huge advantage of getting people, onboarded , getting them productive. It's not a frustrating place to come in, understand things that don't make sense or they're unnatural. So it makes that transition really smooth. And then two, you know, think about the , the diversity of the workforces today and , and all the areas that we have distribution centers. And one of the, the real, I think, game changing features to this is to what Omer talked about, about the, the next bot presenting itself, when it presents itself to you, it can actually sense what language based on your badge that you prefer. And there's, I think a couple dozen different languages, even in slang form of the language , uh, traditional language that , um, that the users can pre-identify with. And then the , the bot is literally speaking their language. So you start talking technology that makes sense in the language that a person is used to. I mean, that , that's huge. Yeah . And , and really thinking, and I've , I've managed some buildings that had very diverse workforces in terms of ethnicities and, and language. And that's really, it's uncomfortable. I mean, you think about yourself if you're in an environment and you're not familiar with the language, it's very uncomfortable. So talk about being the employer of choice, where you can walk in everything's in, you know, what you're used to. It really is a competitive advantage in , in my eyes. Yeah. Cuz

Speaker 2 23:04

Remind me, you said in your , one of your facilities there , what's the number of languages spoken

Speaker 3 23:09

Here ? This was in my , uh, past life. Okay . We had actually, it was here in Orlando though, facility, there were 44 different nationalities in the building. Yeah. And something we really embraced, but it was, but also something from a communication perspective, it certainly provided a lot of challenges. And even within one is a , is a for instance here, there were a lot of , uh, Hispanic cultures in there and there's a lot of variations back when we were talking about even slang within yeah . The Spanish language that locus has went out and, and actually had adopted the slang. So that's talk about unique. Yeah. And really, you know, adapting something to, to meet the needs of the folks. Yeah.

Speaker 2 23:52

Great. Okay. So last question, you've taken on the picking process and it's to both of you, you know, what are you looking at next from an automation standpoint, as you build out this new warehouses and hopefully take on a lot more demand?

Speaker 3 24:06

Sure. Yeah. A couple things we we've already started to explore is , you know , DHL has a great relationship with Boston dynamics and we're looking at testing an , an automated trailer unloader mm-hmm <affirmative>. And um , so we're looking at starting that potentially this summer, as we look at the new warehouses, continue to expand the , the locus application, but also the , the pallet handling , uh, potentially, you know , the put away retrieval replenishment piece. And then also in our design for facility next year, we're also looking at potential some deep palletization to build , to feed a , you know, kind of an ASRS type type setup . So I know , uh, details and obviously in process of testing that and already using that in a lot of the applications. So when the time's right for us to test it out, you know, we're gonna start, but that's, that's the plan to add couple of those. Yeah . For the, for the fit.

Speaker 2 25:04

Yeah. Good. ER,

Speaker 4 25:06

Yeah, I think Tony hit the nail on the head. You know, when we're looking at what's gonna bring us the most benefit, what's gonna help our associates the most on the warehouse floor and make the lives easier. You know, we've addressed the unit picking aspect, which is where a lot of the effort comes and unloading is another one where we've got case level unloading, which is a very difficult task. And you mentioned the partnership with Boston dynamics. We've been working with them for a couple years on that solution, getting it to the point where it's gonna fit this application. And we're really excited about rolling that out. Uh , and then AGVs, as you mentioned at the pallet load, you know, we've been working on that for a few years in different operations as well and have a couple of deployments. And again, I think because we're there, it should be something we're gonna look at with car . Car's a really a great customer from a solution perspective because they bring all the challenges that we traditionally see with different customers under one roof. So we're able to kind of apply a lot of those technologies. And one more, I would say, not to be lost in the discussion is that gives us access to a lot of data. These emerging technologies from the robots, from the warehouse management system or the control system for heavier automated sites. And we're also developing algorithmic optimization within our warehouses. So looking at how we can use that data to better assign tasks to the floor, better optimize our workflow . And I think that's one thing that we don't wanna miss in this conversation is how are we using all that information that's generated from all these technologies, making sure we're then again, improving upon them as we continue to evolve.

Speaker 2 26:35

Yeah. Sounds like maybe the next episode <laugh> we could talk a lot about these things, cuz it's an exciting time in the industry and a lot, lot of stuff happening. So Tony, thanks to you for the partnership. I mean it's exciting to be associated with brands like Carhartt and explore these interesting new spaces in the industry overall and Omer . Thanks to you as always appreciate you guys, both being in here and I thank you . Wish you good times for the rest of this conference and good luck in the , in the future. Thank you. Thank you.

Will Heywood 27:03

If you enjoyed the conversation today, please share it with a friend and rate us on apple podcasts. You can find us online at dhl.com/all business, no boundaries and follow us on LinkedIn and Twitter at DHL supply chain. We'll see you next time.