Automation in the warehouse is progressing at a rapid pace, with robots taking over an increasing number of tasks. The result is speedier and more efficient order fulfillment, at a time when consumers are demanding ever-faster delivery. In this conversation with SupplyChainBrain Editor-in-Chief Bob Bowman, Vince Martinelli, head of product and marketing with RightHand Robotics, reviews the current and future state of robots in the warehouse.

SCB: What kinds of tasks are piece-picking robots performing today?

Martinelli: They’re doing very simple item-handling tasks in warehouses fulfilling e-commerce business. Pharmacies, drug stores, grocery, general merchandise — all of these are using piece-picking robots today. It's an exciting time.

SCB: How sensitive are they? How small an item can they pick?

Martinelli: Our gripper is rated to about two kilograms, and we can grab an item that is up to around 40 centimeters on the long side. Something like 90% of the products and inventory that flow through e-commerce today are less than two kilograms. We’ve picked things as small as packages of salt, lipstick and things like that. The fingers gently cradle the item and help to hold it against suction so it doesn’t get dropped.

SCB: Is a single robot going around picking all the contents of a given order, then bringing them to the packaging station?

Martinelli: No. The robot is generally stationary. When an order is placed, the inventory tote comes out and presents itself. The robot has a camera that allows it to see into the tote. Generally the tote is holding a single SKU, but there could be a bunch of items mixed in.

SCB: How much area can a single fixed robot cover within a given aisle or set of racks?

Martinelli: The robot is typically positioned at the head of the row or aisle, and some goods-to-person mechanism is bringing the inventory to it. Its range is limited by the reach of the arm — you're looking at about a meter and a half. You can reach one or two source totes with it.

SCB: Can this type of robot trace its ancestry back to the factory floor?

Martinelli: Yes, in a way. Robot arms have been around for 30 to 50 years. They're made in high volumes and have gotten very reliable, with options between cobots and more traditional models. So there are plenty of arms to choose from. We consciously said, "We're not going to invent a new arm, but we’re going to focus on the parts that are new and required." In a factory system, the environment and tasks are structured in a way to make it easy for the robot. It doesn't need to see items or make choices about which ones to pick. It’s designed to place items in exactly the same position and orientation, and it’s always going to go to the same location where it sets them down.

In a warehouse, however, there might be 25 items in a bin, all lying at different angles and in slightly different locations within the tote. Each time the robot arm picks one up, it moves differently. Each pick is unique. Now throw in the fact that there might be 100,000 products over a year, and the vision skills and artificial intelligence of the machine has to be entirely different from a factory robot. That’s the challenge.

SCB: A lot of warehouses use robots as conveyors, to direct a human to the pick face. You've got a robot that does the actual picking. Is this technology eventually going to take humans out of the warehouse completely?

Martinelli: Let me break that into several parts. I was part of Kiva Systems — I joined in 2007, when we were just starting out with our first customer project. So I do have familiarity with the other type of warehouse robots. Those systems are great for improving efficiency of the pickers who would have been walking up the isles. They let warehouses use existing static racks, where they would have had human pickers walking around. Our system marries those types of systems. After a person put the items in the tote, it goes downstream, where the items eventually get sorted into orders. Now what does this mean for people in the warehouse? There are several trends at work at the same time. One is the growth of online ordering, even before COVID-19, which forced everyone to move more quickly in areas like grocery home delivery. E-commerce jumped ahead probably five years from what people were forecasting. This has put a stress on warehouse operators, especially when it comes to the availability of labor. It’s making them more productive, helping to support the growth of their business.

Whether it’s our system or anyone else’s, there’s still a need for technicians to look after the robots and monitor the systems. Maybe there aren’t as many people as there were doing the item-handling job, but we're talking about automating the simplest task. It’s not a job that people are excited about or look forward to. I sometimes compare it to switchboard operators from the 1950s or '60s. There are a lot fewer of them today, but the number of jobs in general has grown, so that people can enjoy more challenging tasks with greater variety.

SCB: What are some additional benefits of this technology?

Martinelli: It’s about how you eliminate friction — the ideal state of how, when you order something, it comes out of the factory and lands on your doorstep. I think automation gets you closer to that. You can streamline and get leaner inventories, where you don’t have to hold as much in warehouses. You might be able to afford smaller fulfillment centers that are closer to customers. Automation opens up opportunities to change your whole flow of distribution. It’s going to be interesting to see over the next 10 to 20 years how people solve that puzzle.