Fast-growing My Jewellery deliberately opts for step-by-step automation

How do you automate the warehouse of a company growing 25 to 30 per cent a year? To answer that question, Dutch consultancy Groenewout, part of EPG Consulting, organised a tour of My Jewellery’s warehouse. This fast-growing jewellery and fashion brand has deliberately chosen step-by-step automation with a starring role for nearly 90 autonomous mobile robots. ”Most companies start with automation of order picking, but the biggest savings are actually in the packaging process.”

By Marcel te Lindert

In 2011, Sharon Hilgers started selling jewellery via Facebook. Less than 15 years later, My Jewellery operates both online and offline in the Netherlands, Belgium, Germany, France and England. It doesn’t stop there. ”We are busy adding several languages to our webshop. In addition, we now have forty of our own shops,” says Suzanne Hendriks (pictured), supply chain manager at My Jewellery. ”The third important sales channel is business-to-business. We mainly use this to increase our brand awareness outside the Netherlands. For instance, we sell our products on Transavia planes and at Eindhoven and Düsseldorf airports. We are also now supplying a large department store in Copenhagen to investigate whether our formula is catching on in Scandinavia.”

All sales channels are supplied from My Warehouse, the nearly 10,000-square-metre distribution centre that opened its doors in the Dutch city of Den Bosch two years ago. It is not only the rapid growth of 25 to 30 per cent a year that is demanding on the operation, but also My Jewellery’s strategy. ”We spot trends, buy a product once and sell it out as quickly as possible. That way, we receive 30 to 40 new products in our warehouse every week. Those items account for 60 per cent of the volume. The rest consists of never-out-of-stock products,” Hendriks explains. ”Besides facilitating growth, we are constantly innovating, including in the area of warehouse automation.”

Automation in steps

My Jewellery deliberately chose to keep the operation in its own hands. ”We believe that the growth we are experiencing and the different sales channels we serve are very challenging. And that outsourcing this to a logistics provider only adds complexity,” explains Hendriks, who stresses that My Jewellery cannot do everything itself. ”When I joined four years ago, the company already had big plans for automation and mechanisation. Because I had no knowledge of that, we had to hire external expertise.”

That expertise was provided by Groenewout. Together, they decided on step-by-step automation of the warehouse. ”Because the old system had its limitations, it was clear from the outset that we first had to select and implement a new Warehouse Management System (WMS). That is the reason why, with the move to the new warehouse, we did not start automating immediately, but continued with the existing manual process first,” reports Étienne Teunissen, logistics consultant at Groenewout.

Packing first, then picking

And another striking decision is the choice to start automating the packing process instead of the order picking process. ”Many companies start with order picking because that requires the most people, but it is often in packing that the biggest savings can be made. When automating the packaging process, we see very short payback periods of one to two years, while those for an automatic order picking system quickly go towards five or six years,” Teunissen explains. ”The cost of packaging materials is often lower with automatic packaging than with manual packaging. But the main saving concerns labour costs. On a packing table, an employee might pack 40 orders per hour, while a machine achieves a capacity of 900 to 1,000 boxes per hour.”

Popular in fulfilment centres are wrap around machines, where a box is made to measure and folded around the products. In My Warehouse, they opted for a mix of box erectors and height-reducing carton closing machines. ”This choice is largely to do with My Jewellery’s products. The jewellery is often only on a small card in smooth foil packaging. The chances of losing them in a wrap around machine were high. A concept with box erection machines has the advantage that we can pick directly in the shipping box.”

Mobile robots from Geek+

For automation of the order picking process, My Jewellery chose a system with mobile robots from Geek+. This system has 21 tall robots that do nothing but pick stock totes from the 8-metre-high racks and drop them off at the bottom location. There, they are picked up by one of over 66 small and manoeuvrable robots that take the totes to one of seven workstations. There, order pickers are ready to pick the right products from the bin and distribute them across the prepared shipping boxes. ”This system has been in operation since January, but we are still in the start-up phase. The formal handover of the system has yet to take place,” says Teunissen.

Groenewout examined several concepts, from Autostore to shuttle system. ”Autostore was ruled out because of the inbound and outbound capacity required. My Jewellery needs to get so many order lines out of the system that the Autostore robots would get in each other’s way,” Teunissen explains. Another option is the system where mobile robots bring a complete shelving unit to a workstation. ”An interesting concept, as we can put products that are often sold together in one cabinet and therefore handle several order lines at once. The disadvantage is the limited height, so we would need a lot of space or would have to work with mezzanines.’

Flexible and scalable

A shuttle system was not considered because of My Jewellery’s rapid growth. ”In a feasibility study, the question always remains whether My Jewellery is able to maintain that growth of 25 to 30 per cent per year. What if market conditions are disappointing and growth stagnates?” asks Teunissen. ”A solution with mobile robots is very scalable. We can increase capacity every year as needed by deploying extra robots or adding extra racks. If growth stagnates at some point and expansion is not necessary, the system still continues to pay for itself.”

That is different with an inflexible shuttle system. Because interim expansion is expensive, My Jewellery should have built a shuttle system with sufficient capacity for several years. If then, after, say, two years, growth stagnates, the savings on operational costs calculated for that year would disappear. ”So then the pre-calculated payback period is no longer achieved. That is an important reason for opting for a concept with mobile robots,” Teunissen explains.

Future investments

While the Geek+ system is not yet fully used, Groenewout and My Jewellery are already looking ahead. What are the options to further automate the process and increase capacity? ”We can scale the Geek+ system, but we can also think of other solutions, such as placing an order picking robot on a workstation. We can also increase capacity by switching to batch picking. This allows us to get more order lines from the presented totes, but then we need a pouch sorter or sorting robots to distribute the picked items back to the robots.”

For each solution, Groenewout calculated from which year – taking into account the growth projections until 2030 – implementation becomes interesting. ”This shows that an investment in batch picking only becomes interesting at the end of that horizon. For the time being, with the current volumes it is more cost-efficient to pick directly in the shipping boxes and not to build in an extra sorting layer. Purchasing an order picking robot, on the other hand, is already interesting now. We have enough volume to equip one workstation with it and thus save one order picker per shift.”

The post Fast-growing My Jewellery deliberately opts for step-by-step automation appeared first on Supply Chain Movement.