Meeting Mim
Editor James Burman visits BotsandUs, a robotics company based in London. Andrei Danescu and Oana Jinga, two of the founders of the business, explain about how these fully autonomous robots collect real time data and digitise spaces.
Oana Jinga (OJ): Our product is called Mim, a robot that automates two essential processes within the warehousing environment. Number one is acceptance. Mim will go around any pallets that are being unloaded from trucks and is able to capture the dimensions and condition of the goods, to see if anything has been damaged or altered. The information is then integrated either into a WMS or created as a report that can be passed to other relevant teams, like those in charge of revenue and compliance, and so on.
The second part of it is automating the stocktaking processes. Usually, warehouses have a team of people that walk between the shelves and try and capture the location of every pallet or every box on the shelf. Ascertaining how many there are, if they’ve been moved, or if there’s an empty space instead of a pallet; Mimwill automate all of that. You can get a perfect reconciliation between the digital data that you have in the WMS and what’s actually happening on the warehouse floor at any given time.
JB: What kind of results would you expect undertaking these operations, compared to a human running the same task?
OJ: Primarily, if you look at a direct comparison with a human being, we’re expecting a much faster process of capturing the data – something like 20-30 times faster. The robot only needs one drive by the rack; the human would have to potentially get a scissor lift and go up and down to scan all the boxes. Secondly, you can build up a full 3D digital version of a warehouse, which a human being can’t necessarily do.
JB: How do they charge, and how is power managed?
Andrei Danescu (AD): The robots themselves are battery powered. When they run low on battery power, they will go to one of the charging docks and recharge themselves. The robot management system, or the distributed fleet systems that we have, looks at the number of tasks available for the entire fleet of robots, and then asks how much can this specific robot charge itself before it needs to go back out and continue its mission? This way there’s no downtime or interruption in the data insights feeding into the WMS.
JB: What about maintenance?
AD: Like any electromechanical system, there is a fair amount of wear and tear is associated with it. We’ve engineered and built our robots and systems to ensure the robots’ sensors allow for predictive maintenance, more than reactive maintenance. We know at any point in time what the state of the wear is like on any component, which allows us to be a lot more responsive with the way we approach things. Of course, we do actually have to do physical maintenance on the robots eventually. For us, it’s a matter of swapping a robot out with another to ensure the customers can continuously utilise the tech.
JB: How does the robot interact with current operations? I’m thinking WMS and other automation solutions.
AD: The robots are fully autonomous, and there’s no need for any kind of infrastructure change. The only requirement is to find the space to fit the charging station. They are designed for dynamic working environments, so they will operate with and amongst forklifts, pallet jacks, everything. From that point of view, you could look at them as cobots. The system is inherently designed to be safe to be used around people too.
JB: What scales of operation are suitable for this kind of technology? Is there an upper or lower limit? Are there any other limitations?
OJ: There’s not really an upper or lower limit; you can always put more robots in and cover more space. Everything is based on the number of scans. The robots you actually need for the space is dictated by the number of SKUs. When it comes to the kind of spaces that we potentially can’t operate in, sometimes it’s more of an environment challenge. For example, we talked with a couple of customers that have dirtier or messier warehouses because of the nature of like the materials and the things that they carry, and that could be more of an issue for the bots. Lastly, there’s still a number of customers that are heavily paper focused – some of them don’t even have digitised labels on the pallets, so there might be some limitations in terms of adopting digital tools and technology.
JB: What kind of investment is required for the solution?
OJ: We operate on a ‘robot as a service’ business model, which is becoming quite the standard for the industry now. We charge a fee per robot per month, and it covers the hardware units, access to the data, and any kind of maintenance or servicing from our side. This obviously means the cost is very limited to start – reducing the entry barrier for adopting this kind of technology. Then you can scale up and down based on how many units you actually need.
