Collaborating with a third-party logistics management firm and an international material handling specialist, BMW runs a production parts warehouse in South Carolina that operates with the precision of a Z4 roadster.
Peter Bradley is an award-winning career journalist with more than three decades of experience in both newspapers and national business magazines. His credentials include seven years as the transportation and supply chain editor at Purchasing Magazine and six years as the chief editor of Logistics Management.
Any effort to describe the operations of the production parts warehouse at BMW's assembly plant in Spartanburg, S.C., leads inevitably to comparisons with the finely engineered vehicles that emerge from the other end of the process.
BMW has earned a reputation for producing automobiles and sports utility vehicles that are fast, responsive and reliable. In the same way—from the coordination of inbound materials through the material handling process and sequencing of parts and out to delivery of parts to the production line—the warehouse facility is the result of both careful design and a well-oiled operation that ensures that the right parts arrive at the right place with little wasted motion.
What's more,the process is delivering parts not for mass production but for vehicles that are made to order. Each Z4 roadster or X5 SUV that's built at the Spartanburg facility is the result of an individual order configured with options and color schemes for a particular customer. Literally millions of variations are possible, including left- or right-hand drive and manual or automatic transmissions.
It is the challenge of the entire supply chain,and particularly the warehouse operation, to ensure that the right parts arrive for the right car at the right place at the right time. The plant produces 650 cars every day and the odds are that no two of any model are exactly alike.
The warehouse operation is essentially a DC with a single customer but with several delivery points within the assembly plant. Parts come in from suppliers from around the world and are picked for timely delivery to specific locations in the factory.
Many options, one standard
"Our focus is to deliver cars right on time while allowing customers and dealers as much flexibility to change orders as possible," says Manfred Stöger, vice president of logistics and information technology for BMW Manufacturing Corp. The materials and manufacturing systems allow dealers to adjust orders up to six days prior to the start of production. "That has a lot of advantages for customers and dealers," Stöger says, "but it's a huge challenge for the supply chain."
To deliver the enormous variety of vehicles on time and in the right configurations demands a highly automated system and a cadre of sophisticated suppliers and well-integrated systems. "The warehouse could not handle this manually," Stöger says.
The operation is interesting for a number of reasons. First, while BMW owns the warehouse, management of inbound materials and the internal operations is in the hands of a contract logistics company, TNT Logistics, which has more than 500 employees dedicated to BMW operations on site. Second,the highly automated system for storing and retrieving a large portion of the thousands of parts that go into any vehicle, through tight integration with BMW systems and automatic identification technology, enables much of the precision in the operations. The system was designed and installed by the German material handling company Witron, a specialist in building automated distribution centers.
The company selected Witron in part, says Stöger, because it develops and customizes its mechanical parts and software itself, including the warehouse management system used by the warehouse.
Witron CEO Reinhard Boesl says that the planning involved six months of "intensive discussions" with BMW managers in Spartanburg and in Munich. The actual installation took about 18 months. "We had tough timelines," Boesl says . "The roof was not done when we began installing systems." Another few months were required to ramp up and test the system.
Boesl continues, "As soon as we achieved a degree of integration, we invited BMW to do testing. The testing process is a good opportunity to double-check if the detailed specifications match the practical requirements. It's a critical phase. It allows the client to see what he is getting."
The complex system can adapt as BMW makes changes to its operations, though the flexibility is not unlimited.
Responsive handling
On the other hand, the system adapted quickly to increases in production and model changes. Since the automated systems have been up and running, BMW and its partners have made a number of adjustments to the operation. For instance, the company added a third induction point in the warehouse—conveyor locations that move material to storage locations. When BMW ended production of the Z3 roadster and started building the Z4, the system adapted without a single change to software code. In late September, the plant began producing 2004 X5s that included 2,100 new parts—with only a 10-car gap between the old and new models. The warehouse is flowing parts for 600 to 650 cars a day with a system initially designed to support production of 450. Stöger says, "We'll be able to increase production to 700 units without additional storage space."
About half the content of the vehicles flows through the warehouse, which handles about 13,000 active parts for the Z4 and X5. A second warehouse within a few miles of the main facility handles hazardous materials and the largest and bulkiest parts. Those include any materials larger than a pallet and any that are mo re than 2.5 meters high.
Ronald Feld, section manager for physical logistics for BMW, probably has the most day-to-day familiarity with the warehouse operations. Because each car is somewhat different from every other car, getting the right parts to the line is perhaps the central mission of the warehouse. "I'm sequencing single parts," Feld says. But the system is efficient enough that it requires notice only three to four hours before parts are needed on the assembly line. "When a car enters the assembly area from the stacker, its [identification number] generates a broadcast signal to us in the warehouse and to direct sequence suppliers." The system also calculates the expected time to deliver the materials.
BMW's automated warehouse
Pallet warehouse
Eight AS/RS pallet cranes
Picking left or right five pallets deep
38,800 cubic meters storage volume
17,500 pallet storage locations
928 sequence picking locations
Tray warehouse
Five AS/RS tray cranes
Picking left or right one deep
23,500 tray locations
3,900 cubic meters storage volume
Daily activity
2,200 pallet receipts
2,500 tray receipts
2,200 outbound pallets via conveyor
3,300 tray movements, in and out
8,000 sequence picks
Source: Witron
Dual systems
To keep operations running smoothly, Witron installed both pallet and tray automated storage and retrieval systems. The pallet warehouse is a high-bay facility with 17,400 storage locations on six storage levels. The Witron system has a number of fail-safe parts to its design, Feld explains. For example, pallets are stacked six wide at each level. Should a crane fail, the system allows other cranes to retrieve pallets five deep, allowing complete redundancy if required. In addition, the system is designed such that if the system wants a pallet that is three deep, for example, the cranes can store the first two in different areas to retrieve the third, all the while keeping track of the first t wo. "The warehouse reorganizes itself I don't know how many times a day," Feld says . To enable that, the system reserves 15 percent of the storage locations to allow it to relocate pallets.
The warehouse is not fully automated, however. A substantial amount of picking occurs by hand. Employees pick and scan each part. "It's all picks of one," Feld explains. So while a tote containing visors contains enough for 80 vehicles, each one is destined for a specific car. When a worker completes a pick in one location, he or she receives automated notification of his or her next stop. "This optimizes the operation," Feld says. "[A worker follows] the shortest path with the fewest steps."
The automated tray warehouse has 26,000 storage locations, and each storage location can hold up to 10 part numbers. The t ray handling system can handle individual loads of up to 220 pounds, compared to 2,200 pounds for the pallet handling system. Further, the tray system is simpler than the pallet system: Trays are stored one deep on each side of the large racks, and the automated picking equipment need only select to the left and right.
The system provides protection against breakdowns by storing each part number in more than one tray and in separate aisles, so if one machine breaks down, parts are still accessible for other parts of the warehouse. About 70 percent of the storage locations are in use.
Trays selected by the automated system move to one of five identical workstations. Four of the locations are used for picking and putaway, while the fifth is reserved for unusual requirements. Even that is part of the efficiency plan: Keeping the fifth position in reserve helps compress the travel time for the automatic guided vehicle that transports material to and from the work stations. Typically, one worker can man two of the stations.
Inbound controls
On the inbound side, materials are scanned on arrival and data uploaded via an RF system. Data capture then occurs several other times as goods move to putaway locations and eventually to staging for the shop floor. After shipments are received, palletized loads move to one of two major induction points, or i-points as they are called by BMW. "A scan marries the container to a bar-coded pallet," Feld explains. The pallets then move via an automated system to the pallet warehouse. As the pallets move down the conveyor, automated devices automatically check for contours,height and weight and divert any pallets that might need any correction, a process handled manually.
A large portion of the inbound shipments are in returnable containers that will be sent back to suppliers for reuse. "We promote returnables," says Feld. "We're making giant steps with our European suppliers. They were shipping 100- percent corrugated. Now we're getting 30- to 40-percent returnables. Our NAFTA suppliers are shipping 95-percent returnables." Those suppliers, he reports, only ship in disposable corrugated containers when they are short of returnables.
The actual material handling in the warehouse is overseen by BMW, but operations are in the hands of TNT Logistics North America, the lead logistics provider for the company. TNT manages both inbound transportation and material processing and sequencing in the warehouse. It processes more than 10,000 inbound and outbound shipments each day: the outbound sequencing, which involves 28 parts families, is managed through Witron's warehouse management system.
TNT, BMW suppliers and BMW operations are closely integrated from origin until delivery to the production line. Deb Hall, vice president of l ogistics solutions for TNT North America , summarizes the process: "BMW provides the demand file, which provides all information on material from each supply point. We schedule by optimizing loads and provide that to suppliers."
Road masters
Carriers are responsible for signing off on shipments picked up at the supplier locations. Both the carrier and the supplier have to agree on the contents of a shipment: Drivers report any discrepancies to TNT's Jacksonville call center.While physical receiving takes place at the warehouse, BMW has immediate visibility into any exceptions.
Currently, five truckload carriers and one LTL carrier bring in the bulk of the shipments.TNT also operates a small dedicated fleet performing a few milk runs, shuttles between warehouses and truckload shipping. Some 425 trucks a day move through the warehouse bays over two shifts. Those loads are destined either for the plant or the warehouse. Once the materials have arrived, TNT continues to manage the process for BMW, managing the yard and staging of trailers. Shipments are scanned and received at the dock and validated against what's expected. Goods then move either directly to the production line or to the warehouse. (BMW manages ocean and air freight transportation. It consolidates shipments out of Europe in Germany to maximize weight and cube utilization, sending about 9,000 40-foot containers a year to the South Carolina facility.)
Hall says, "A lot of 3PLs manage transportation. The difference here is that we are actually managing materials. BMW does not view the route or the driver—it doesn't need to see that. BMW personnel see things that are of interest to them. They can see exceptions. There's a proactive notification, so there's no monitoring needed by BMW."
She adds, "BMW is a very demanding customer, but it doesn't expect more than it should." She meets with Stöger every month to gain an executive level view of operations. In addition, the TNT operations managers on site meet with BMW's materials and logistics managers daily. "Our staff and their staff are integrated," Hall says. "They are very forthcoming.
They want us involved in the planning process. … We were fortunate to find a customer that would allow us to be integrated, not only in transportation but in material planning. They allowed us to step over a line we're not allowed to step over very often."
Progress in generative AI (GenAI) is poised to impact business procurement processes through advancements in three areas—agentic reasoning, multimodality, and AI agents—according to Gartner Inc.
Those functions will redefine how procurement operates and significantly impact the agendas of chief procurement officers (CPOs). And 72% of procurement leaders are already prioritizing the integration of GenAI into their strategies, thus highlighting the recognition of its potential to drive significant improvements in efficiency and effectiveness, Gartner found in a survey conducted in July, 2024, with 258 global respondents.
Gartner defined the new functions as follows:
Agentic reasoning in GenAI allows for advanced decision-making processes that mimic human-like cognition. This capability will enable procurement functions to leverage GenAI to analyze complex scenarios and make informed decisions with greater accuracy and speed.
Multimodality refers to the ability of GenAI to process and integrate multiple forms of data, such as text, images, and audio. This will make GenAI more intuitively consumable to users and enhance procurement's ability to gather and analyze diverse information sources, leading to more comprehensive insights and better-informed strategies.
AI agents are autonomous systems that can perform tasks and make decisions on behalf of human operators. In procurement, these agents will automate procurement tasks and activities, freeing up human resources to focus on strategic initiatives, complex problem-solving and edge cases.
As CPOs look to maximize the value of GenAI in procurement, the study recommended three starting points: double down on data governance, develop and incorporate privacy standards into contracts, and increase procurement thresholds.
“These advancements will usher procurement into an era where the distance between ideas, insights, and actions will shorten rapidly,” Ryan Polk, senior director analyst in Gartner’s Supply Chain practice, said in a release. "Procurement leaders who build their foundation now through a focus on data quality, privacy and risk management have the potential to reap new levels of productivity and strategic value from the technology."
Businesses are cautiously optimistic as peak holiday shipping season draws near, with many anticipating year-over-year sales increases as they continue to battle challenging supply chain conditions.
That’s according to the DHL 2024 Peak Season Shipping Survey, released today by express shipping service provider DHL Express U.S. The company surveyed small and medium-sized enterprises (SMEs) to gauge their holiday business outlook compared to last year and found that a mix of optimism and “strategic caution” prevail ahead of this year’s peak.
Nearly half (48%) of the SMEs surveyed said they expect higher holiday sales compared to 2023, while 44% said they expect sales to remain on par with last year, and just 8% said they foresee a decline. Respondents said the main challenges to hitting those goals are supply chain problems (35%), inflation and fluctuating consumer demand (34%), staffing (16%), and inventory challenges (14%).
But respondents said they have strategies in place to tackle those issues. Many said they began preparing for holiday season earlier this year—with 45% saying they started planning in Q2 or earlier, up from 39% last year. Other strategies include expanding into international markets (35%) and leveraging holiday discounts (32%).
Sixty percent of respondents said they will prioritize personalized customer service as a way to enhance customer interactions and loyalty this year. Still others said they will invest in enhanced web and mobile experiences (23%) and eco-friendly practices (13%) to draw customers this holiday season.
That challenge is one of the reasons that fewer shoppers overall are satisfied with their shopping experiences lately, Lincolnshire, Illinois-based Zebra said in its “17th Annual Global Shopper Study.”th Annual Global Shopper Study.” While 85% of shoppers last year were satisfied with both the in-store and online experiences, only 81% in 2024 are satisfied with the in-store experience and just 79% with online shopping.
In response, most retailers (78%) say they are investing in technology tools that can help both frontline workers and those watching operations from behind the scenes to minimize theft and loss, Zebra said.
Just 38% of retailers currently use AI-based prescriptive analytics for loss prevention, but a much larger 50% say they plan to use it in the next 1-3 years. That was followed by self-checkout cameras and sensors (45%), computer vision (46%), and RFID tags and readers (42%) that are planned for use within the next three years, specifically for loss prevention.
Those strategies could help improve the brick and mortar shopping experience, since 78% of shoppers say it’s annoying when products are locked up or secured within cases. Adding to that frustration is that it’s hard to find an associate while shopping in stores these days, according to 70% of consumers. In response, some just walk out; one in five shoppers has left a store without getting what they needed because a retail associate wasn’t available to help, an increase over the past two years.
The survey also identified additional frustrations faced by retailers and associates:
challenges with offering easy options for click-and-collect or returns, despite high shopper demand for them
the struggle to confirm current inventory and pricing
lingering labor shortages and increasing loss incidents, even as shoppers return to stores
“Many retailers are laying the groundwork to build a modern store experience,” Matt Guiste, Global Retail Technology Strategist, Zebra Technologies, said in a release. “They are investing in mobile and intelligent automation technologies to help inform operational decisions and enable associates to do the things that keep shoppers happy.”
The survey was administered online by Azure Knowledge Corporation and included 4,200 adult shoppers (age 18+), decision-makers, and associates, who replied to questions about the topics of shopper experience, device and technology usage, and delivery and fulfillment in store and online.
An eight-year veteran of the Georgia company, Hakala will begin his new role on January 1, when the current CEO, Tero Peltomäki, will retire after a long and noteworthy career, continuing as a member of the board of directors, Cimcorp said.
According to Hakala, automation is an inevitable course in Cimcorp’s core sectors, and the company’s end-to-end capabilities will be crucial for clients’ success. In the past, both the tire and grocery retail industries have automated individual machines and parts of their operations. In recent years, automation has spread throughout the facilities, as companies want to be able to see their entire operation with one look, utilize analytics, optimize processes, and lead with data.
“Cimcorp has always grown by starting small in the new business segments. We’ve created one solution first, and as we’ve gained more knowledge of our clients’ challenges, we have been able to expand,” Hakala said in a release. “In every phase, we aim to bring our experience to the table and even challenge the client’s initial perspective. We are interested in what our client does and how it could be done better and more efficiently.”
Although many shoppers will
return to physical stores this holiday season, online shopping remains a driving force behind peak-season shipping challenges, especially when it comes to the last mile. Consumers still want fast, free shipping if they can get it—without any delays or disruptions to their holiday deliveries.
One disruptor that gets a lot of headlines this time of year is package theft—committed by so-called “porch pirates.” These are thieves who snatch parcels from front stairs, side porches, and driveways in neighborhoods across the country. The problem adds up to billions of dollars in stolen merchandise each year—not to mention headaches for shippers, parcel delivery companies, and, of course, consumers.
Given the scope of the problem, it’s no wonder online shoppers are worried about it—especially during holiday season. In its annual report on package theft trends, released in October, the
security-focused research and product review firm Security.org found that:
17% of Americans had a package stolen in the past three months, with the typical stolen parcel worth about $50. Some 44% said they’d had a package taken at some point in their life.
Package thieves poached more than $8 billion in merchandise over the past year.
18% of adults said they’d had a package stolen that contained a gift for someone else.
Ahead of the holiday season, 88% of adults said they were worried about theft of online purchases, with more than a quarter saying they were “extremely” or “very” concerned.
But it doesn’t have to be that way. There are some low-tech steps consumers can take to help guard against porch piracy along with some high-tech logistics-focused innovations in the pipeline that can protect deliveries in the last mile. First, some common-sense advice on avoiding package theft from the Security.org research:
Install a doorbell camera, which is a relatively low-cost deterrent.
Bring packages inside promptly or arrange to have them delivered to a secure location if no one will be at home.
Consider using click-and-collect options when possible.
If the retailer allows you to specify delivery-time windows, consider doing so to avoid having packages sit outside for extended periods.
These steps may sound basic, but they are by no means a given: Fewer than half of Americans consider the timing of deliveries, less than a third have a doorbell camera, and nearly one-fifth take no precautions to prevent package theft, according to the research.
Tech vendors are stepping up to help. One example is
Arrive AI, which develops smart mailboxes for last-mile delivery and pickup. The company says its Mailbox-as-a-Service (MaaS) platform will revolutionize the last mile by building a network of parcel-storage boxes that can be accessed by people, drones, or robots. In a nutshell: Packages are placed into a weatherproof box via drone, robot, driverless carrier, or traditional delivery method—and no one other than the rightful owner can access it.
Although the platform is still in development, the company already offers solutions for business clients looking to secure high-value deliveries and sensitive shipments. The health-care industry is one example: Arrive AI offers secure drone delivery of medical supplies, prescriptions, lab samples, and the like to hospitals and other health-care facilities. The platform provides real-time tracking, chain-of-custody controls, and theft-prevention features. Arrive is conducting short-term deployments between logistics companies and health-care partners now, according to a company spokesperson.
The MaaS solution has a pretty high cool factor. And the common-sense best practices just seem like solid advice. Maybe combining both is the key to a more secure last mile—during peak shipping season and throughout the year as well.