Robert Stobiński, Amica

I've been working in industry, logistics, distribution, and production for over 25 years. I had very interesting challenges and worked with Amazon's robots for the last 5 years, and I especially fondly remember launching the robot center in London — it was a very, very interesting 18 months. I'll share a few observations about this high-bay warehouse as well. I'd like to say fewer words myself and instead show two interesting short films, a few minutes each. The first will describe the warehouse from the user's perspective but also from the perspective of the companies that participated in this investment project. Then I'd like to show a film that doesn't focus on the challenges of building the warehouse but rather the challenges the city faced when the components and elements needed to build it had to pass through it.

Alright, let's begin. I hope the Amica brand is not entirely foreign to you. Indeed, this company has existed for over 60 years in its current form, and over the years it has tried to find its place on the global market of appliance manufacturers. For the time being, we sell our products in over 60 countries. We have a concept of focusing on brands known locally — hence in Poland it's Amica, but for example in Scandinavian markets it's Gram, in the Russian market it's Hansa, and in the English market it's CDA. And recently, a few weeks ago, we signed an agreement with the owner of the Fagor brand in Spain and undertook an attempt to revitalize that brand — we believe it has great potential. In terms of product types, we produce both large appliances — meaning cookers, ovens, and refrigerators — we also offer refrigerators, washing machines, and dishwashers, but also small appliances like irons, vacuum cleaners, blenders, and other products of that type.

This is what our high-bay warehouse looks like. Here are some logistical data — I'll talk about it in a moment. It is certainly the tallest building in its region, and as far as high-bay warehouses go, at least at the time we checked, there was no taller one in Poland. But before we get to what it looks like and how it works, here is a schematic overview of the interior. I'd like to share some of the challenges, because it's difficult to do something that nobody before us has really done. When we look for a solution and find a so-called best practice, we go to a manufacturer and say: 'Look, we want what you built there or what another company built for another client.' In our case, we weren't able to find any practical, working solution in our industry that worked exactly like we needed. We had other ideas, including static storage where we knew the stackers worked, but in our case unfortunately it didn't. This made the normal investment process extremely difficult, because together with the suppliers we had to learn everything. And because there was no experience from an industry perspective, the process of selecting the final solution supplier was also not easy — it required many trials and errors.

The second thing — which maybe didn't emerge right at the beginning during implementation but became very clear before the warehouse launch — is the answer to the question of whether the system managing this warehouse's operations should be a local system developed by the company that supplies the stackers, or whether it should be the IT system normally used in the enterprise. Amica uses the SAP program. Through various discussions, comparing benefits and risks, we made a joint decision that we believe the warehouse manufacturer would have greater competence in managing the warehouse than the software manufacturer, and we decided to use the manufacturer's system. To this day, we have never once regretted this decision. I can't answer the question of whether we would also have been satisfied had we decided to use the SAP program for managing the movement of individual products or pallets. I simply wanted to share the experience that you need to answer this question somewhere at the beginning, because of course if you choose your own software — independent of your existing systems — it creates a huge topic: you have to build an interface, and that's not an easy process. It requires many iterations, many tests, and you have to be prepared that at the beginning things won't necessarily run smoothly.

The industry has very visible sales seasonality, with sales in the fourth quarter being significantly higher. On the other hand, in summer there is one product category that sells very well — refrigerators. Because when there's a threat of food spoiling, we want to deal with it quickly and buy the product. On the other hand, imagine the investment cycle for huge facilities in Poland's geography and weather — if we take into account that the investment cycle from start of construction to completion takes a little over 12 months, then if we don't plan around winter, we end up finishing construction exactly at the moment when our peak sales season begins. The question then becomes: what to do — open or wait? I think each company should answer that question independently, and I can accept the assumption that there will be different answers and each will be correct.

In our warehouse we can store almost 300,000 products. They don't come from production in a single day. If we assume we can produce, let's say, 10,000 units per day, then with 300,000 units and production of 10,000, it would take us a month to fill the warehouse. On the other hand, if before the warehouse launch we have these products stored somewhere else — I mean in another external warehouse — then we'd be delivering goods from two warehouses, which creates an entirely separate project related to the relocation of current logistics resources. Do we want to do this all at once, or do we want to start selling from the new warehouse only from a certain point? Do we want to deliver products from two places for a period of time? Or do we want to first transfer everything, and only after completing that process deliver from the previous warehouse, then launch the new one, and only upon launching the new warehouse close the previous one by transferring the last batches of products?

And finally, the last key issue — partly discussed during our previous panel — how to build the project team. Should it be an independent team that works only within this single project, or should it be a team composed of people who normally perform other duties but devote part of their time to a standalone project? I apologize for not giving specific answers about what we chose at Amica, because my goal was more to show the challenges that every company facing an investment task must address. As I said, this isn't mathematics where we know that two plus two equals four — although there are probably different theories about whether it's really four or a bit more or less. But there isn't one correct answer to these different challenges. The answer is situational and depends on each company's circumstances.

Alright, here is basically what I mentioned before — the warehouse has a height of 27 meters, a capacity of 26,000 pallet places. I think this is one of the interesting changes in Amica's logistics system compared to most likely the rest of the appliance industry, where products are stored in blocks and each truck used to retrieve products has a special gripper. Depending on the logistics system the manufacturer uses, these grippers can capture, I don't know, four or eight or some other number of products at once. Unfortunately, at Amica this can't be done — there are no boxes, only pallets. Therefore, we had to develop a plan for which products, in what quantities, and on which pallets we want to store. A hint from me: there isn't one pallet size — there are different pallet dimensions with different capacities.

As for the throughput of this warehouse, we can issue over 180 pallets per hour and receive about 120 pallets per hour. If we multiply this by the number of products we can fit on a pallet, we get several thousand units per hour. To date, we've been using this warehouse for about two years, and there hasn't been a single moment where the warehouse couldn't send products to the picking zone or receive products from production. Rather, we have trouble managing the picking zone — the warehouse still has better efficiency than we do when retrieving products from it.

What is very interesting, as I also mentioned during the panel, is that we didn't design this warehouse just to have an automated warehouse. If any of you have ever been to the city of Września, you probably know that in a relatively small area are located Poland's two largest home appliance manufacturers — as Wojtek mentioned, Poland is the center of the appliance industry in Europe — namely Samsung and Amica. There simply isn't room, and to make matters worse, Amica's facilities in Września are located between a prison building and Samsung. Amica can develop only toward the city, but it has no possibility of acquiring additional land. We wanted to have a warehouse, so our first question was: what must we do to have many products on a small area? And the answer was obvious: we must stack them vertically.

Our previous storage system allowed us to stack at most 12 cookers one on top of another and about 6 refrigerators, and that was all. Increasing the height created safety risks for operators, so that solution was unacceptable. That's how the idea for a high-bay warehouse was born, which as you can see here increased our storage area by 7.5 times. Many different technical factors were used to make this warehouse a reality. As I mentioned, we handle about 60 full truckloads per day — about 1,600 products. And what's very interesting: if we consider who manages the warehouse's operations, it's only one person per shift. In our case, working three shifts, that means three people operate this warehouse in a single day.

Today we conduct most deliveries between our warehouse and our customers' warehouses using full new vehicles, so it's relatively easy and fast to fill a truck and send it to our customer. But we increasingly notice that in this race to shorten supply chains, business customers would increasingly like to deliver products directly from the manufacturer to the end consumer, bypassing their own warehouse. This means that if, for example, I can have 80 cookers in one truck going to one business customer — easy, fast, and pleasant — but suddenly if each of those 80 wanted us to send them to 80 different customers in 80 different locations across the country, instead of one truck I'd probably have several. I don't want to say 80 because they could be consolidated somehow, but it certainly wouldn't be as easy and pleasant. History teaches that it's not what we want but what the market decides, so we're thinking about preparing for this new challenge and being able to deliver products individually directly to consumers.

A very interesting solution is the visualization of all warehouse operations, which uses the simple light system — green, yellow, red — operated by the one person managing the warehouse. With this system, she can immediately answer whether everything is working correctly or whether there are any doubts about whether a given stacker is operating. If it's red, we can directly contact the people who can first check the communication — if it works correctly and there are no application issues — and then proceed to the more difficult task of checking the mechanical part. This is particularly difficult because there are no corridors, only narrow passages for the stackers, which creates enormous challenges for someone who has to enter at the 47th level and check something. We have special procedures and equipment for handling such cases.

We talked today about Industry 4.0 — we could also talk about Logistics 4.0. In this specific case, we don't use all available techniques — I mean, for example, augmented reality has absolutely no application here, neither does 3D printing. We also don't have simulations of how this warehouse operates on a daily basis — though of course before starting the investment we did many such simulations. What we do have is something that is extremely critical and that companies sometimes either forget or underestimate — namely cybersecurity. Today, similar to banks, if we saw a shoplifting incident, that alone isn't dangerous. What's dangerous are coordinated attacks prepared over several months, often taking over user accounts and only striking after a period of time. These attacks paralyze a company's operations. We can periodically read about such cases — recently there was the well-known case of Maersk, which faced the dilemma of either paying a ransom or ceasing to operate entirely, because the penetration was so deep there was no way out without help. So I encourage everyone, especially when using various software and cloud services, to remember to protect yourself well.

To summarize: what is undoubtedly the strong side of an automated warehouse is the very small footprint relative to how many products we can load. Because it's automated, it generates far fewer errors — hundreds of times fewer than human operators. This is also very important, because when we confirm that something can be sent to a customer, it actually exists in the warehouse and we know exactly where it is. The algorithm automatically enforces FIFO — first in, first out. The rotation in the warehouse is under absolute control, which is particularly important not only in our industry but also for products with expiration dates. If there are any questions, I'll be happy to answer them now.

The supplier of this warehouse happened to be a foreign company, but it wasn't their foreign origin that decided the choice — what mattered was their experience with similar solutions.