The staff at J. Jill's DC couldn't say enough about the speed and accuracy of RF-based packing systems ? until they saw what could be done with lights.
James Cooke is a principal analyst with Nucleus Research in Boston, covering supply chain planning software. He was previously the editor of CSCMP?s Supply Chain Quarterly and a staff writer for DC Velocity.
Last December, the staff at J. Jill's Tilton, N.H., DC finally saw the light—or in this case, the lights: 356 illuminated LED displays. After years of using a radio-frequency (RF)-based order fulfillment system, the facility was abandoning its handheld scanners in favor of something a bit flashier.
J. Jill, which sells women's apparel, shoes, and accessories, wasn't just updating its look, however. It was making the switch to accommodate growth. Originally a catalog business, the retailer has shifted its focus to online and retail sales. Since it opened its first store in 1999, it has pursued an aggressive expansion strategy, adding roughly 40 stores annually in the last few years. Its DC "network," however, has undergone no such expansion. The 425,000-square-foot facility in Tilton continues to supply all of J. Jill's stores nationwide (which currently number 231) in addition to filling online and catalog orders.
As throughput volume grew, it became clear that the DC would have to replace its old packing system with something speedier. "We had a decent rate with the RF," says Glenn Broderick, J. Jill's director of retail distribution. "Order fillers were doing 575 units an hour with 99.95 percent accuracy." But that wouldn't be enough to keep up with the company's aggressive growth plans, he explains. "We were looking to drive the throughput north of 800 units an hour."
The search for a replacement technology that wouldn't require significant modification to the material handling equipment (or extensive worker retraining) led J. Jill to light-directed order fulfillment technology. But instead of the more familiar pick-to-light technology, the retailer opted for a variation: a put-to-light system.
Put-to-light systems work in much the same way as pick-to-light systems, but in reverse. With pick-to-light systems, illuminated displays are mounted at stock items' storage locations to indicate to pickers which items to retrieve. With put-to-light systems, the light displays are instead mounted at packing stations equipped with stationary cartons used to collect items for individual orders. Instead of telling pickers which items to retrieve, the flashing lights indicate where to distribute them.
Late last year, J. Jill installed AL Systems' DynaPack putto-light system, a solution that incorporated 356 light display modules managed by a scalable software platform. Though the company expected to see a jump in productivity, Broderick says, it was still caught off guard by the size of the gains—and by the speed with which they were achieved.
Age of enlightenment
Light-directed order fulfillment technology is hardly new. The first pick-to-light systems date back to the '80s. Rapistan (since renamed Dematic) is generally credited with being the first U.S. material handling equipment vendor to offer the technology.
In the early days, the systems' use was pretty much limited to the cosmetic and pharmaceutical industries, where the need for order picking accuracy justified the cost. "They were used where products were high priced, consumer service was a high priority, and mistakes [in order picking] very expensive," says Stephen Small, vice president of marketing and sales for Lighthouse Selection, a maker of pick-to-light equipment in Manchester, N.H.
Before long, users noticed that the systems did more than just boost accuracy. They also boosted productivity, largely because workers no longer had to stop to consult pick tickets or scanner displays for directions. As word of the productivity benefits got out, the systems caught on with a broader base of users. Among them were a number of large retail chains, which began installing pickto- light systems during the '90s.
The technology is best suited to operations that require high-velocity picking of a limited number of stock-keeping units (SKUs). But it's important to note that those fastmoving items must be stored next to a conveyor pick line. "If you have to do a lot of walking for an order, you're not going to buy pick to light. You'll never get payback on it," says Steve Mulaik, a consultant with the firm Progress Group in Atlanta. "Successful implementation of pick to light takes place where confirmation and search time [exceed] the walk time."
In the last five years, voice-directed systems have emerged to challenge light systems for the order selection technology market. But pick to light still has the edge in certain applications, including high-decibel operations. "In a noisy environment, lights work much better than voice," says Jack Kuchta, executive vice president at the warehouse consulting firm Gross & Associates in Woodbridge, N.J. Stephen Small estimates the current market for pick-to-light systems at somewhere between $40 million and $50 million a year.
Rapid advances
Pick-to-light technology has come a long way since its introduction. One result has been that light systems have become much cheaper and easier to install than they were in the past. A decade ago, outfitting a DC with a pick-tolight system meant physically hand-wiring a display module to every workstation in the network. The advent of socalled "bus technology" for electronic devices in the last five years changed all that. Bus technology standardizes datainterface hookups for electrical devices (the USB and FireWire technologies familiar to personal computer users are examples of bus technology).
"With bus technology, you don't have to hard-wire each display," says Tom Singer, a principal at the consulting firm Tompkins Associates, based in Raleigh, N.C. "You lay out a bus frame and it lowers the installation costs. It also makes replacement of the units more cost effective."
Integration costs have also dropped in recent years. As part of the installation process, a pick-tolight system's software must be linked with the user's warehouse management system (WMS), which feeds order information to the light system. In the past, that often meant costly custom programming work. Today, however, many vendors supply either pre-built interfaces or toolkits designed to make it easy to create the necessary interfaces. That's made light systems eminently more marketable, says Kuchta. "When you have to build the interfaces, it made the decision to install one more precarious," he says. "It's the interfaces that drive up costs."
Despite the advances in software and bus technology, taking the pick-to-light route still requires a hefty investment. Singer says a company can expect to spend between $100 and $125 per light display. In addition, it should figure on spending another 50 percent for integration, even with the availability of pre-built interfaces. "If you have 10,000 lights, I'd figure $1.5 million for the system," says Singer.
The variation known as the put-to-light system, however, gets around that problem. With put to light, you don't need a module to identify the location of each SKU; you just need one for each order collection point. A DC that fills orders for 100 stores from a stock of 15,000 SKUs would need only 100 display modules, not 15,000.
Fast and accurate
Put-to-light technology is particularly well suited to specialty retailers that push a limited number of SKUs out to their stores, says Christopher Castaldi, a vice president of client development at AL Systems Inc. Like pick-to-light systems, put-to-light systems are designed for "dense picking" operations, where the SKUs needed for order fulfillment can be stored near the packing station.
J. Jill's operation is a case in point. Though the retailer's New Hampshire distribution center holds 38,000 SKUs, only 14,000 of those SKUs are supplied to stores. That makes it feasible to position these items close to the conveyors— a factor Broderick considers key to the system's success. "The put to light works for us because of the density of our picks," he says.
Installation was a simple matter of mounting the display modules onto existing two-tiered shelving— each module identifies the pack location assigned to an individual store. Though the company installed 356 display modules, only 231 are currently in use. That leaves more than 100 modules to accommodate growth.
Each pack station is located off a conveyor branch from the sortation system. At the start of the conveyor line, a worker takes a case of product from storage (each case contains a single SKU) or from a recently arrived inbound shipment that has been cross-docked. About 80 percent of J. Jill's merchandise is made overseas, primarily in Asia. The retailer brings the containers into the country through the Port of Long Beach, moves them cross country via rail, and finally trucks them to New Hampshire. Outbound orders move to its stores via UPS and to online and catalog customers with DHL.
Once the worker has selected a case, he or she breaks out the items—say, shoes—and puts them into a tote. Then the worker rolls the totes, two at a time, down a skatewheel conveyor located beside the put-to-light stations.
The worker first takes items out of one tote, depositing them into the cartons for various stores according to the instructions displayed on the put-to-light modules. The worker then makes a downstream pass, filling orders from that tote, and then returns upstream to his starting point, taking items out of the second tote. This approach allows the worker to distribute two totes' worth of a product in a single pass.
The new process has also cut back on the need for scanning. Before the put-to-light system was introduced, the order fulfillment process included two scanning steps. Workers first scanned each incoming carton of merchandise in order to get their packing directions, which popped up on the scanner's screen. Then, after depositing an item into a carton, they had to scan that carton again for verification purposes.
With the put-to-light system, by contrast, the worker scans the carton just once, at the beginning of the process. Almost instantly, the light modules at the appropriate packing stations begin to flash, displaying the required product quantities. Because there's no further need for scanning, workers don't have to take the scanners with them on their rounds anymore. "Now you're pretty much hands-free," says Broderick, "because you're not carrying that RF [scanner gun] around."
On the face of things, the changes brought by the new light system might seem insignificant. But taken together, they've done a lot to rev up the operation. For example, eliminating the need to perform a verification RF scan has cut a whole step out of the packing process, speeding up order fulfillment with no sacrifice in accuracy. On top of that, the DC has found that workers are more productive now that an associate can empty two totes in one swing down the conveyor line through the pack zone.
Although J. Jill knew fulfillment rates would rise, it didn't realize how big the gains would be. "It almost doubled what we were expecting in productivity," says Broderick. The company was looking for a gain in the 20-percent range, he explains; but after two months, it was clear that the new light system had actually boosted efficiency by a whopping 40 percent.
The New York-based industrial artificial intelligence (AI) provider Augury has raised $75 million for its process optimization tools for manufacturers, in a deal that values the company at more than $1 billion, the firm said today.
According to Augury, its goal is deliver a new generation of AI solutions that provide the accuracy and reliability manufacturers need to make AI a trusted partner in every phase of the manufacturing process.
The “series F” venture capital round was led by Lightrock, with participation from several of Augury’s existing investors; Insight Partners, Eclipse, and Qumra Capital as well as Schneider Electric Ventures and Qualcomm Ventures. In addition to securing the new funding, Augury also said it has added Elan Greenberg as Chief Operating Officer.
“Augury is at the forefront of digitalizing equipment maintenance with AI-driven solutions that enhance cost efficiency, sustainability performance, and energy savings,” Ashish (Ash) Puri, Partner at Lightrock, said in a release. “Their predictive maintenance technology, boasting 99.9% failure detection accuracy and a 5-20x ROI when deployed at scale, significantly reduces downtime and energy consumption for its blue-chip clients globally, offering a compelling value proposition.”
The money supports the firm’s approach of "Hybrid Autonomous Mobile Robotics (Hybrid AMRs)," which integrate the intelligence of "Autonomous Mobile Robots (AMRs)" with the precision and structure of "Automated Guided Vehicles (AGVs)."
According to Anscer, it supports the acceleration to Industry 4.0 by ensuring that its autonomous solutions seamlessly integrate with customers’ existing infrastructures to help transform material handling and warehouse automation.
Leading the new U.S. office will be Mark Messina, who was named this week as Anscer’s Managing Director & CEO, Americas. He has been tasked with leading the firm’s expansion by bringing its automation solutions to industries such as manufacturing, logistics, retail, food & beverage, and third-party logistics (3PL).
Supply chains continue to deal with a growing volume of returns following the holiday peak season, and 2024 was no exception. Recent survey data from product information management technology company Akeneo showed that 65% of shoppers made holiday returns this year, with most reporting that their experience played a large role in their reason for doing so.
The survey—which included information from more than 1,000 U.S. consumers gathered in January—provides insight into the main reasons consumers return products, generational differences in return and online shopping behaviors, and the steadily growing influence that sustainability has on consumers.
Among the results, 62% of consumers said that having more accurate product information upfront would reduce their likelihood of making a return, and 59% said they had made a return specifically because the online product description was misleading or inaccurate.
And when it comes to making those returns, 65% of respondents said they would prefer to return in-store, if possible, followed by 22% who said they prefer to ship products back.
“This indicates that consumers are gravitating toward the most sustainable option by reducing additional shipping,” the survey authors said in a statement announcing the findings, adding that 68% of respondents said they are aware of the environmental impact of returns, and 39% said the environmental impact factors into their decision to make a return or exchange.
The authors also said that investing in the product experience and providing reliable product data can help brands reduce returns, increase loyalty, and provide the best customer experience possible alongside profitability.
When asked what products they return the most, 60% of respondents said clothing items. Sizing issues were the number one reason for those returns (58%) followed by conflicting or lack of customer reviews (35%). In addition, 34% cited misleading product images and 29% pointed to inaccurate product information online as reasons for returning items.
More than 60% of respondents said that having more reliable information would reduce the likelihood of making a return.
“Whether customers are shopping directly from a brand website or on the hundreds of e-commerce marketplaces available today [such as Amazon, Walmart, etc.] the product experience must remain consistent, complete and accurate to instill brand trust and loyalty,” the authors said.
When you get the chance to automate your distribution center, take it.
That's exactly what leaders at interior design house
Thibaut Design did when they relocated operations from two New Jersey distribution centers (DCs) into a single facility in Charlotte, North Carolina, in 2019. Moving to an "empty shell of a building," as Thibaut's Michael Fechter describes it, was the perfect time to switch from a manual picking system to an automated one—in this case, one that would be driven by voice-directed technology.
"We were 100% paper-based picking in New Jersey," Fechter, the company's vice president of distribution and technology, explained in a
case study published by Voxware last year. "We knew there was a need for automation, and when we moved to Charlotte, we wanted to implement that technology."
Fechter cites Voxware's promise of simple and easy integration, configuration, use, and training as some of the key reasons Thibaut's leaders chose the system. Since implementing the voice technology, the company has streamlined its fulfillment process and can onboard and cross-train warehouse employees in a fraction of the time it used to take back in New Jersey.
And the results speak for themselves.
"We've seen incredible gains [from a] productivity standpoint," Fechter reports. "A 50% increase from pre-implementation to today."
THE NEED FOR SPEED
Thibaut was founded in 1886 and is the oldest operating wallpaper company in the United States, according to Fechter. The company works with a global network of designers, shipping samples of wallpaper and fabrics around the world.
For the design house's warehouse associates, picking, packing, and shipping thousands of samples every day was a cumbersome, labor-intensive process—and one that was prone to inaccuracy. With its paper-based picking system, mispicks were common—Fechter cites a 2% to 5% mispick rate—which necessitated stationing an extra associate at each pack station to check that orders were accurate before they left the facility.
All that has changed since implementing Voxware's Voice Management Suite (VMS) at the Charlotte DC. The system automates the workflow and guides associates through the picking process via a headset, using voice commands. The hands-free, eyes-free solution allows workers to focus on locating and selecting the right item, with no paper-based lists to check or written instructions to follow.
Thibaut also uses the tech provider's analytics tool, VoxPilot, to monitor work progress, check orders, and keep track of incoming work—managers can see what orders are open, what's in process, and what's completed for the day, for example. And it uses VoxTempo, the system's natural language voice recognition (NLVR) solution, to streamline training. The intuitive app whittles training time down to minutes and gets associates up and working fast—and Thibaut hitting minimum productivity targets within hours, according to Fechter.
EXPECTED RESULTS REALIZED
Key benefits of the project include a reduction in mispicks—which have dropped to zero—and the elimination of those extra quality-control measures Thibaut needed in the New Jersey DCs.
"We've gotten to the point where we don't even measure mispicks today—because there are none," Fechter said in the case study. "Having an extra person at a pack station to [check] every order before we pack [it]—that's been eliminated. Not only is the pick right the first time, but [the order] also gets packed and shipped faster than ever before."
The system has increased inventory accuracy as well. According to Fechter, it's now "well over 99.9%."
IT projects can be daunting, especially when the project involves upgrading a warehouse management system (WMS) to support an expansive network of warehousing and logistics facilities. Global third-party logistics service provider (3PL) CJ Logistics experienced this first-hand recently, embarking on a WMS selection process that would both upgrade performance and enhance security for its U.S. business network.
The company was operating on three different platforms across more than 35 warehouse facilities and wanted to pare that down to help standardize operations, optimize costs, and make it easier to scale the business, according to CIO Sean Moore.
Moore and his team started the WMS selection process in late 2023, working with supply chain consulting firm Alpine Supply Chain Solutions to identify challenges, needs, and goals, and then to select and implement the new WMS. Roughly a year later, the 3PL was up and running on a system from Körber Supply Chain—and planning for growth.
SECURING A NEW SOLUTION
Leaders from both companies explain that a robust WMS is crucial for a 3PL's success, as it acts as a centralized platform that allows seamless coordination of activities such as inventory management, order fulfillment, and transportation planning. The right solution allows the company to optimize warehouse operations by automating tasks, managing inventory levels, and ensuring efficient space utilization while helping to boost order processing volumes, reduce errors, and cut operational costs.
CJ Logistics had another key criterion: ensuring data security for its wide and varied array of clients, many of whom rely on the 3PL to fill e-commerce orders for consumers. Those clients wanted assurance that consumers' personally identifying information—including names, addresses, and phone numbers—was protected against cybersecurity breeches when flowing through the 3PL's system. For CJ Logistics, that meant finding a WMS provider whose software was certified to the appropriate security standards.
"That's becoming [an assurance] that our customers want to see," Moore explains, adding that many customers wanted to know that CJ Logistics' systems were SOC 2 compliant, meaning they had met a standard developed by the American Institute of CPAs for protecting sensitive customer data from unauthorized access, security incidents, and other vulnerabilities. "Everybody wants that level of security. So you want to make sure the system is secure … and not susceptible to ransomware.
"It was a critical requirement for us."
That security requirement was a key consideration during all phases of the WMS selection process, according to Michael Wohlwend, managing principal at Alpine Supply Chain Solutions.
"It was in the RFP [request for proposal], then in demo, [and] then once we got to the vendor of choice, we had a deep-dive discovery call to understand what [security] they have in place and their plan moving forward," he explains.
Ultimately, CJ Logistics implemented Körber's Warehouse Advantage, a cloud-based system designed for multiclient operations that supports all of the 3PL's needs, including its security requirements.
GOING LIVE
When it came time to implement the software, Moore and his team chose to start with a brand-new cold chain facility that the 3PL was building in Gainesville, Georgia. The 270,000-square-foot facility opened this past November and immediately went live running on the Körber WMS.
Moore and Wohlwend explain that both the nature of the cold chain business and the greenfield construction made the facility the perfect place to launch the new software: CJ Logistics would be adding customers at a staggered rate, expanding its cold storage presence in the Southeast and capitalizing on the location's proximity to major highways and railways. The facility is also adjacent to the future Northeast Georgia Inland Port, which will provide a direct link to the Port of Savannah.
"We signed a 15-year lease for the building," Moore says. "When you sign a long-term lease … you want your future-state software in place. That was one of the key [reasons] we started there.
"Also, this facility was going to bring on one customer after another at a metered rate. So [there was] some risk reduction as well."
Wohlwend adds: "The facility plus risk reduction plus the new business [element]—all made it a good starting point."
The early benefits of the WMS include ease of use and easy onboarding of clients, according to Moore, who says the plan is to convert additional CJ Logistics facilities to the new system in 2025.
"The software is very easy to use … our employees are saying they really like the user interface and that you can find information very easily," Moore says, touting the partnership with Alpine and Körber as key to making the project a success. "We are on deck to add at least four facilities at a minimum [this year]."