Robotics & Warehouse Automation July 2019

Order Fulfilment Mobile Robots Start to Deliver – more than 500,000 to be installed in next 5 years

Ash Sharma
Ash Sharma

Ash is the MD of IA and lead for our Robotics and Warehouse Automation Division. He brings 20 years of experience to the table in sectors ranging from industrial automation and smart manufacturing to drones, robotics and medical technology.

The acquisition of Kiva Systems by Amazon and its subsequent dominance of online retailing through heavy adoption of robotics and automation has triggered two significant trends in the industry. First, it led to the emergence of dozens of robotics start-ups to fill the vacuum that Kiva left behind; and second, it has forced retailers and logistics companies to adopt automation to keep pace with Amazon. Consequently, we believe that we’re fast approaching the tipping point for autonomous mobile robots (AMRs) used to support order fulfilment. Excluding Amazon, we predict that more than 100,000 AMRs will have been deployed for order fulfilment by the end of 2020, and that more than 580,000 will be installed over the next five years (See fig 1)

Delivery Drivers

Although fewer than 300 sites had order fulfilment AMRs deployed at the end of last year, explosive growth is predicted due to the ‘perfect storm’ of a number of key drivers:

The Amazon Effect. Retailers, ecommerce companies and 3PLs are being forced to employ order fulfilment AMRs in order to provide the speed, flexibility and low-cost delivery that Amazon has pioneered, whilst at the same time mitigating the issue of a labor shortage.
Rapid growth of ecommerce. The rapid and consistent growth of ecommerce is forcing retailers to make substantial changes to their supply chains and fulfilment strategies. The vast number of customer deliveries that need to happen daily (and importantly at peak times) has created a huge burden on 3PLs and retailers. Either more human workers have to be employed, or robots.
Labor shortage. Record low unemployment rates in major economies, coupled with the on-going shift of younger generations away from manual labor has created a major shortage of warehouse workers.
Early adopters have proven technology. Over the last 3-4 years, early adopters of order fulfilment AMRs have been the guinea pigs with their pilots. However, those pilots have now turned to full-scale deployments of hundreds and in some cases, thousands of robots. New customers now typically have an accelerated deployment with limited piloting.
RaaS & lease models. The offering of a low CAPEX approach to deploying order fulfilment AMRs has opened the technology up to a much larger TAM and reduced the risk for the customer.

Investment in AMR Vendors Pours in

Throughout 2018 AMR companies accelerated the pace at which they raised funding. Close to $400m was raised in 2018, most notably by GreyOrange ($140m) and Geek+ ($150m). Geek+ also announced that it had raised a further $150m earlier this month – just 8 months after its last round of funding. (We note that it’s not clear at this point if the full $150m has been raised, or just a portion). Geek+ installed more than 7,000 AMRs in 2018 and according to our analysis was the largest vendor with a share of over 25% of the market. Its strong growth last year was mirrored by its domestic counterparts and the three largest vendors were all Chinese.

One important question that could be asked is, if Geek+ is growing so rapidly and shipping so many robots, then why is it burning through so much cash? This can be explained by the fact that although the company has shipped and installed several thousand AMRs, many of these have been on a lease basis, or are used within its own warehouses. Many of its peers are using the same RaaS/leasing approach, which whilst being very beneficial for getting volumes deployed, is creating challenges in financial stability with revenues being pushed out.

Goods-to-Person vs. Person-to-Goods vs. Piece Picking?

With Amazon deploying Kiva’s goods-to-person approach for order fulfilment, its easy to see why others like GreyOrange, Geek+ and Quicktron have followed in its footsteps, and arguably have seen greatest success so far. Despite this, AMR vendors have emerged with a variety of different approaches and sub-approaches. Person-to-goods AMRs and mobile piece and unit picking approaches all have their own merits, and drawbacks. The question over which technology or approach will ‘win’ is not an easy one to answer. All approaches are forecast for strong growth (albeit at different paces and timeframes), and the approach preferred will depend on the customer type, SKU count, the type of product being handled, throughput, budget, warehouse type and size and more factors. Whilst goods-to-person AMRs are the preferred solution for large retailers, they may not be deployed in major quantities at 3PLs that operate on shorter contracts and with slimmer margins.

Whilst in every industry there are always winners and losers, right now it appears that the size of the untapped opportunity and the underlying drivers of demand are more than strong enough to support all the vendor types we see today, and as a result we expect very promising growth for the majority.


For more information on our Mobile Robots insights, please contact us at