The initial take up of warehouse automation has been limited due to nervousness around the changing market and the risks associated with investment. Many retailers have dabbled with put walls, pick to light technology and a more traditional MHE approach but there is still a long way to go in terms of full warehouse automation.
Despite the hype, warehouse bots are not yet the norm. At the moment, at the most cutting edge of technology are small robots that follow a light to pick up products in the DC. This kind of technology is being used by some retailers and will of course suit certain retailers better than others. But despite the innovation the costs of deploying such robotics in DCs are still significant, so whilst the technology is available, it isn’t necessarily being deployed en masse.
Five year forecast
There was a rush to automate around five years ago before a subsequent lull, but now retailers are refocusing their efforts on automation and all that comes with it. There might not be a move to advanced robotics for any retailer just yet, but there will be a take up in more traditional automation methods across the retail environment. It is also important to remember that current robotics may work well for retailers that only stock 1,000 SKUs, but fast fashion retailers for example can have more than ten times that. Designing robots to accommodate that volume would be very difficult.
One of the main challenges is actually a physical constraint. Not all warehouses are a perfect shape – there are a huge number of warehouses that are converted mills, for example, that have multiple floors or various sections. So, in this scenario it would be difficult for a robot to navigate and efficiently pick items. Secondly, the nature of products differs dramatically from retailer to retailer. It would very difficult at the current stage of technology to program a robot to deal with the mass variety of products. And let’s not forget that the more advanced the technology becomes, the more expensive it will be. This is a huge constraint for any retailer looking to implement new technology in the store or the warehouse.
Another constraint robot developers will face will be the amount of items per order. The current average of items per order is 1.8 which is not too difficult for a robot to comprehend. But once you move into grocery retail or fast fashion, the items per order will be significantly higher than this and can cause some confusion for an automated bot.
Automation also poses some customer experience issues, for retailers that may sell luxury goods for example, as it lacks the personal touch. The idea of it being hand-picked and packaged gives it a bespoke element, and this will be a customer expectation. However, we may begin to see a combination of automation in this environment, where the item may be picked by a robot but checked and packaged by a worker.
With regard to full warehouse automation, retailers are concerned that these huge investments will not pay off, so the investment is currently piecemeal as opposed to automating the whole warehouse. But eCommerce is here to stay and organisations are beginning to recognise the need to automate and we will see the implementation of new technologies accelerate over the coming years.
Read the full article about warehouse automation featuring Manhattan Associates on Raconteur.