December 14, 2019

Robotics are changing ecommerce fulfillment - DigitalCommerce360

Ecommerce distribution centers look markedly different than they did just a few years ago. A growing number of retailers are leveraging picking, packing and delivery robots to automate key tasks—and the shift may be just getting started.

More than 50,000 warehouses worldwide are expected to include commercial robotics by 2025. That would be a dramatic jump from the 4,000 in existence in 2018, according to market research firm ABI Research. In the United States alone, there are expected to be roughly 23,000 robot-powered warehouses in operation by 2025, up from 2,500 today in 2018.

Ecommerce’s growing market share—and the push to deliver consumers’ online orders the same day or next day after a shopper places an order—is helping drive that growth, says Nick Finill, senior analyst at ABI. Retailers’ growing needs also help explain why the number of fulfillment-related jobs jumped 41% between 2018 and 2015 to 1.2 million warehousing and storage jobs from 785,000 in 2015, according to the U.S. Bureau of Labor Statistics. That growth has come amid a tightening labor market as the U.S. unemployment rate has fallen to 3.6% in October from 5.5% in January 2015.

Those conditions have driven retailers to deploy the latest generation of sophisticated robotics. Those robots can handle a range of tasks—from autonomous mobile robots that can transport inventory to articulated robotic arms that can manipulate items to automated storage and retrieval robots that retrieve items for use. And many are ideally suited to ecommerce fulfillment, Finill says.

“Robotics in the supply chain has the potential to reduce costs, improve efficiencies and increase human productivity and accuracy,” says David Glick, chief technology officer at logistics and fulfillment company Flexe, who previously worked as director of global fulfillment at Inc.


Gap adds robotics to speed fulfillment

Driven by the need for more flexible and efficient ecommerce fulfillment, Gap Inc. recently added a robotic, artificial-intelligence-powered picking system from robotics vendor Kindred.

The apparel company added the system at the same time that it was revamping its fulfillment network to better support ecommerce orders. Rather than overhaul its 14 distribution center network, Gap tested the robotics system in 2017 by integrating it into the existing infrastructure at its Gallatin, Tennessee-based facility, says Kevin Kuntz, the retailer’s senior vice president of global logistics fulfillment.

The robotics technology has helped speed up the facility’s fulfillment operations and helped it weather an influx of orders during peak sales seasons, Kuntz says.

“Robots enable warehouses to scale operations up or down as required while offering major efficiency gains and mitigating inherent challenges associated with labor and staffing,” ABI’s Finill says.

The technology Gap deployed, Kindred’s SORT system, transformed the retailer’s processes. Gap previously dropped thousands of products into a pile from a giant oval belt—known as a Bombay sorter—into manual sorting stations, and an employee would pick up each piece and scan it with a bar code scanner. A corresponding light on the putwall—a vertical grid of cubbies—would then light up to signal an employee which cubby corresponds to that order. After all items were scanned, the employee took the contents of the cubby and passed it off to another station to be packed and shipped.


The SORT system enables Gap to reconfigure the chutes from the Bombay sorter to send products directly to the robotic arm sorter. The system uses a camera to look at the items in the bin, comparing each to all the images of Gap items in its system. It then uses the system’s various data points to calculate and execute an optimal pick strategy for each task in real time, says Kindred CEO Jim Leifer.

For example, if there is a pile of plastic bags with a red T-shirt, black socks and pair of jeans, the arm can look at this unstructured pile of items and determine the right correct piece to select. The arm then grasps, pinches or suctions the item, then swings to expose the item’s bar code to be scanned by one of four cameras surrounding the bin to ensure it matches the order.

Occasionally, the robotic arm encounters an issue and needs human intervention. “While the solution is autonomous, there will be moments where it doesn’t have the confidence level to execute its tasks,” Leifer says. “We have a pilot, so if a robot needs help, then it opens a channel to one of our remote pilots within less than a second and helps the robot understand and keep working. “

Within a few months, the SORT system was picking 98% of Gap’s items autonomously and rarely required pilot assistance. Because of SORT’s success at the Tennessee facility, Gap implemented the system into its Fresno, California, distribution center last fall and plans to add it to its Fishkill, New York, distribution center by the end of this year, the retailer says.

“Kindred enhanced the grippers and scanners to improve the accuracy of the machines,” Kuntz says. “We have seen impressive speed improvements in the 18 months since we began implementation, and the percent of autonomous picks have increased.”


While Kindred declined to reveal the cost of implementing the system, Leifer refers its approach as “robotics-as-a-service” (RaaS), which means the retailer pays per successful pick (it declined to share whether there were any additional fees). By 2026, there are expected to be 1.3 million installations of RaaS by various robotics vendors, according to ABI Research.

Although the system is automated, a warehouse employee has to complete the supply chain flow after the robotic arm picks the order. However, the system greatly increases that worker’s efficiency, he says. “A single employee can pack many more orders at a SORT station because the robotic arm is doing most of the work,” he says. “If the employee oversees multiple picking stations, they can be responsible for two to three times as much throughput.”

How Best Buy transformed its warehouse

Robotics have also transformed Best Buy Co. Inc.’s fulfillment processes.

About six years ago, the retailer was struggling to meet consumers’ delivery expectations, and it knew it needed to make significant changes if it was going to survive, says Rob Bass, Best Buy’s chief supply chain officer. It wanted to become more efficient, improve its supply chain and figure out how it could compete in a different way.

Stephanie Crets |

But the retailer had to combat several issues: roughly 60% of Best Buy’s ecommerce units shipped are repack items—which means the retailer removes them from their manufacturer’s packaging and places them in Best Buy-branded packaging. But Best Buy didn’t have the capital to invest in a bigger team to keep up with the ecommerce repack item demand, as well as shoppers’ growing demand for more efficient ecommerce fulfillment and delivery. Plus, most of its distribution centers operated five days a week as opposed to seven days, and its systems were “old enough to smoke and drink in the U.S.,” Bass says.


But it did have a few warehouses in strategically important locations: Compton, California, which is near Los Angeles; Piscataway, New Jersey, which is near New York City; and Chicago. Best Buy sought to put its “big-hitting ecommerce SKUs” inside these buildings and transform them into metro ecommerce centers, Bass says.

Best Buy worked with logistics system integrator Bastian Solutions, coupled with AutoStore—a bin storage system in which bins are stacked vertically in a grid and retrieved by robots that travel on the top layer of the system—to overhaul its distribution centers. The process took about 12 months to implement, but the retailer declined to provide specifics of costs or when it started using the new system.

The retailer now has 30,000 bins and 73 robots at each of its three metro ecommerce centers. Best Buy delivers up to 40% of a store’s inventory to a store from its metro ecommerce centers and segregates store delivery by aisle (the rest of the metro ecommerce center is focused on fulfilling online orders). Bastian and AutoStore outfitted Best Buy’s regional distribution centers—located in San Francisco, Atlanta and Findlay, Ohio—with 150,000 bins and 195 robots.

Since implementing these new systems, Best Buy increased its ability to pack up to 20,000 units per hour. The system has made warehouse workers more efficient, Bass says, noting that they used to walk up to eight miles per day retrieving and packing products. They now stand in place while the bins and robots move around the facilities. That enables them to focus on individual tasks instead of running around the facility.

In turn, the retailer’s shipping is now about two days, a far cry from the six days that it previously took to deliver an online order. “We’re very proud of that so we can match Amazon’s speed of delivery without the Prime member price tag,” Bass says.


Small retailers benefit from robotics

It’s not just the large retail chains that are adding robotics. For example, online apparel, accessories and gifts retailer Marleylilly in June 2018 began working with warehouse robot provider Locus Robotics to help it fulfill and deliver orders quickly and efficiently—especially during peak seasons.

“Our peak seasons are fast,” says Tiffany Stewart, assistant manager of fulfillment at the retailer, declining to provide sales specifics.

Locus Robotics’ system uses algorithms to accept clusters of orders for products in the same warehouse section or aisle and rely on warehouse mapping software to follow an optimal path among merchandise racks. Pick workers in the aisles then read order information on the robots to place the order items in bins that the robots then carry to packing stations, navigating around the facility without running into anything else, such as workers or other robots.

Locus also provided a dashboard in the warehouse so management can monitor how many picks are being done per hour, how many items need to be picked for that day and the number of open orders.

It has 15 robots working within its fulfillment center and, during its peak holiday season that number grows to 25, says Jason Everest, director of production at Marleylilly. Therefore, the retailer can scale up the robotics to keep up with peak season demand without hiring additional warehouse staff.


Within nearly a year and a half of using Locus Robotics, Marleylilly tripled the number of orders it ships per month, while also significantly increasing accuracy and pick rates, although it declined to give further details. “We’re not eliminating jobs, we’re really just increasing the efficiency and optimizing our pick operation,” Everest says. “We want to stay ahead of the curve, ahead of our competitors, and to do that, we need to stay technologically advanced.”

That’s increasingly important within a competitive ecommerce environment.

“Collaborative robots drive productivity increases,” says Locus CEO Rick Faulk. “They ensure that brands are able to meet their fulfillment goals, despite the widespread scarcity of warehouse labor and influx in order volumes.”

[email protected] | @StephCretsDC360

Rate this item
(0 votes)

Leave a comment

Make sure you enter all the required information, indicated by an asterisk (*). HTML code is not allowed.

Latest Tweets

Legislature acted correctly in repealing blue law | Editorial - Bismarck Tribune - Legislature acted correctly in r…
OUS employee receives achievement award | Ohio News - Huntington Herald Dispatch - OUS employee receives achievemen…
Mt. Gox Trustee: Creditors Will Soon Receive Decisions Over Rehabilitation Claims - Cointelegraph - Mt. Gox Trustee…
Follow WDTNN on Twitter