Humanoid robots are increasingly being tested in warehousing and logistics facilities around the world, most notably by retail titan Amazon, but will they be a common sight in the future?
Ash Sharma is the managing director of Interact Analysis and lead for the Robotics and Warehouse Automation Division. He brings 20 years of experience to the table in sectors ranging from industrial automation and smart manufacturing to drones, robotics, and medical technology.
Last October’s announcement by online retail behemoth Amazon that it is testing Agility Robotics’ humanoid robot, Digit, in its warehouses caused a stir in the global media, sparking numerous news articles and debate about the ethics of using robots to replace human workers. But does Amazon’s announcement point to a rapid adoption of bipedal robots in the near future?
The concept of humanoid robots is not new. In fact they date back at least as far as Ancient Greece, with the mechanical Servant of Philon1, a humanoid automaton that could pour and mix wine and water. However, there has been a sharp rise in recent years in the number of companies developing and trialing humanoid robots, particularly within the warehouse sector. Digit, which can grasp and lift objects, is going to start by moving empty tote boxes as part of Amazon’s efforts to automate its warehouse operations. Similarly, Figure and Boston Dynamicsare prototyping humanoid robots for use in distribution centers. Tesla’s Optimus Robot also looks promising for warehouse applications, as it can self-calibrate its arms and legs and has the capability to sort objects fully autonomously. Additionally, the commercial launch of Apollo by tech startup Apptronik is expected to take place in late 2024, and videos have already shown it walking, case picking, palletizing, and unloading trailers.
Yet, in spite of the current chatter around the technology and the flurry of pilots and prototypes, change is unlikely to happen overnight. Most pilot projects take months or even years to reach completion, and rollouts tend to happen in incremental stages. Furthermore, our feeling is that, although we are already starting to see humanoid robots appear in warehouses, some obstacles still exist, particularly in regard to acceptance of humanoid robots by their human co-workers.
Are bipedal robots inevitable?
Warehouse automation solutions in general have been in place for decades, carrying out many physically demanding, menial jobs. For example, automated storage and retrieval systems (AS/RS) are already widely used alongside human workers to deliver much faster rates of order picking and to increase throughput. Furthermore, over the past five years, there has been a steady rise in the development and adoption of mobile automation solutions for the warehouse, such as automated guided vehicles (AGVs) and autonomous mobile robots (AMRs).
This surge in demand has been in part fueled by the ongoing labor and skills shortages blighting the industry. Research by Interact Analysis shows scarcity of labor remains the biggest driver of demand for mobile robots and the impact of shortages is becoming more acute. As a result of this and other drivers for mobile automation (such as increasing labor costs, e-commerce growth, and the shift to flexible manufacturing), Interact Analysis forecasts shipments of mobile robots will continue growing at an annual rate of approximately 50% until 2027.2 (See Figure 1.)
Similar to AMRs and AS/RS, humanoid robots may simply be another evolutionary stage in the development of technology for the industry. Because of their ability to move and interact with their environment in a similar way to an actual person, humanoid robots have the potential to meet some very specific needs of the modern warehouse while also offering a very different value proposition to traditional robots. Warehousing involves repetitive and physically demanding menial work that often has seasonal peaks in demand. This work often involves interacting with a variety of different products that lack standardization and a uniform shape and size. While traditional robots are very good at doing the same repetitive work over and over again, humanoid robots can be more adaptable (in a similar way to humans) and therefore can be applied to multiple different tasks throughout a warehouse.
Early applications for bipedal robots are likely to include trailer unloading, which is simple, physically demanding, and repetitive, but difficult to complete using traditional robots. While they tend to operate at a slower pace than traditional robots, humanoid robots also offer the potential to be introduced to the workforce during peak periods without requiring substantial operational changes to warehouse workflows or alterations to the layout of the warehouse. In this way, humanoid robots offer greater flexibility than other materials handling solutions, as they can be dropped into existing supply chains alongside human workers. Humanoid robots are able to take on jobs that are unappealing and take a toll on the human body (such as trailer loading/unloading), freeing up employees for more complex, less physically demanding, and more interesting tasks. Additionally, unlike traditional robots, humanoid robots possess a level of mobility and dexterity that allows them to take on multiple different tasks across facilities, workflows, and applications, and to handle the variety of objects found in a typical warehouse.
Given these potential benefits, will humanoid robots see the same sort of growth rate as AMRs and other robotics solutions? That depends on how well they are able to overcome the barriers to adoption. The largest barrier is the high cost of humanoid robots, which means businesses will currently have to wait a long time to achieve a return on their investment. However, another significant barrier is the “uncanny valley” effect, or the feeling of unease or revulsion people feel when they encounter a human-like robot, and the personification of role replacement humanoid robots represent.
Is our unease surmountable?
At the time of Amazon’s announcement, concerns were raised about humanoid robots displacing human workers.3 Similar concerns have been raised in the past about other robotic technology, such as AMRs. But anecdotal evidence indicates people like working alongside AMRs, where the robot carries out menial, physically demanding work, while they act in a supervisory capacity.
However, the very reasons humanoid robots are capable of working so well alongside human workers are also one of the biggest stumbling blocks for their rapid and widespread adoption: their ability to move and function similarly to a human. From the automaton in ETA Hoffman’s nihilistic 1815 short story “The Sandman” to movies and TV shows such as Terminator, Avengers: Age of Ultron, Blade Runner, and Westworld, humanoid robots have been depicted as problematic and, in some cases, apocalyptically so. Coupled with this, humanoid robots appear to cause greater resentment than other forms of autonomous mobile robots because they are role replacement personified. After all, few humans want to be outpaced, outlifted, and outperformed by a robot that looks like a person.
Responding to concerns about job losses, Amazon has emphasized the “hundreds of thousands of new jobs” that have been created as a result of its use of robotic systems, including “700 categories of new job types in skilled roles,” with robots being used to replace the most “menial, mundane, and repetitive” tasks.
Ultimately, companies are unlikely to be deterred from deploying humanoid robots by their appearance. Although Digit walks on two legs and is capable of lifting and moving objects with its arms, we are far from a dystopic future in which sentient robots blend seamlessly into the human population. The bipedal robot has been designed specifically for warehouse automation where the focus is on increasing throughput and filling labor gaps, rather than the complexities of human thought and movement. Indeed, Amazon describes Digit as “a mobile manipulator solution,” and Tye Brady, chief technologist at Amazon Robotics, told reporters in Seattle that people are “irreplaceable” to the company because of their “ability to think at a higher level, the ability to diagnose problems.”
Promising but still some way off
At Interact Analysis, we have charted the steep rise in demand for warehouse automation technology. Many facilities are still operating manually, but companies worldwide plan to increase investment in automation over the coming years. Skills and labor shortages are showing no signs of stopping within the materials handling industry, and mobile robots are being utilized already in a range of different settings. They often provide a solution to repetitive, physically demanding, uncomfortable, and dangerous jobs.
In addition to this, our research indicates the scalability and flexibility to use humanoid robots within existing warehouse operations alongside human workers could provide a unique answer to ongoing skills and labor gaps. There will always be inertia to change regardless of what the automation solution looks like, and it is too early to tell whether in the long-term bipedal robots will become widely used in warehouses. However, it is certainly a possibility. It will be dependent on the success of early pilots, whether ethical concerns can be overcome, and whether other robotics technology is found to be better suited to specific tasks. Amazon has always been a leader in its use of robotics, with the rest of the industry tending to follow (or fail!), so this pilot could be the catalyst for the wider rollout of humanoid robots in the future. Although competition to develop affordable and effective models is growing, the widespread use of humanoid robots in warehousing, if it happens, is clearly some way off.
J.B. Hunt President and CEO Shelley Simpson answers a question from the audience at the Tuesday afternoon keynote session at CSCMP's EDGE Conference. CSCMP President and CEO Mark Baxa listens attentively to her response.
Most of the time when CEOs present at an industry conference, they like to talk about their companies’ success stories. Not J.B. Hunt’s Shelley Simpson. Speaking today at the Council of Supply Chain Management Professionals’ (CSCMP) annual EDGE Conference, the trucking company’s president and CEO led with a story about a time that the company lost a major customer.
According to Simpson, the company had a customer of their dedicated contract business in 2001 that was consistently making late shipments with no lead time. “We were working like crazy to try to satisfy them, and lost their business,” Simpson said.
When the team at J.B. Hunt later met with the customer’s chief supply chain officer, they related all they had been doing for the company. “We told him that we were literally sitting our drivers and our trucks just for you, just to cover your shipments,” Simpson said. “And he said to us, ‘You never shared everything you were doing for us.’”
Out of that experience, came J.B. Hunt’s Customer Value Delivery framework. This framework, according to Simpson, provides a roadmap for creating value and anticipating customer needs.
Framework for Excellence
J.B. Hunt created the above framework to help them formulate better relationships with customers.
The framework consists of five steps:
Understand customer needs: It all starts, according to Simpson, with building a strong relationship with the customer and then using the information gained from those discussions to build a custom plan for the customer.
Deliver expectations: This step involves delivering on the promises made in that custom plan.
Measure results: J.B. Hunt believes that they are not done when freight makes it to the destination. They also need to measure how successful they were versus what the customer expected from them.
Communicate performance: This step involves a two-way exchange, where J.B. Hunt walks the customer through their performance and gets verbal agreement on whether or not they have met the customer’s needs.
Anticipate new value: Here J.B. Hunt looks at what they are hearing from their customer today and then uses that information to derive what the customer may be looking for in the future.
Simpson said the most important part of the process is the fourth step, communicating performance (perhaps reflecting the piece that went wrong in that initial failed customer relationship).
Not only can this framework be used to drive excellence in a company, but it can also be adapted as a model for driving personal excellence, Simpson said. Instead of understanding the customer needs, the process starts with understanding yourself: what your strengths and interests are. This understanding helps drive a personal development plan and personal goals for the year, which can be measured and assessed. For example, each year, Simpson gives herself a letter grade on each of her personal goals and communicates her assessment back to her boss. She has also found it helpful to anticipate where opportunities lie beyond what she is personally doing.
Confronted with the closed ports, most companies can either route their imports to standard East Coast destinations and wait for the strike to clear, or else re-route those containers to West Coast sites, incurring a three week delay for extra sailing time plus another week required to truck those goods back east, Ron said in an interview at the Council of Supply Chain Management Professionals (CSCMP)’s EDGE Conference in Nashville.
However, Uber Freight says its latest platform updates offer a series of mitigation options, including alternative routings, pre-booked allocation and volume during peak season, and providing daily visibility reports on shipments impacted by routings via U.S. east and gulf coast ports. And Ron said the company can also leverage its pool of some 2.3 million truck drivers who have downloaded its smartphone app, targeting them with freight hauling opportunities in the affected regions by pricing those loads “appropriately” through its surge-pricing model.
“If this [strike] continues a month, we will see severe disruptions,” Ron said. “So we can offer them alternatives. We say, if one door is closed, we can open another door? But even with that, there are no magic solutions.”
Turning around a failing warehouse operation demands a similar methodology to how emergency room doctors triage troubled patients at the hospital, a speaker said today in a session at the Council of Supply Chain Management Professionals (CSCMP)’s EDGE Conference in Nashville.
There are many reasons that a warehouse might start to miss its targets, such as a sudden volume increase or a new IT system implementation gone wrong, said Adri McCaskill, general manager for iPlan’s Warehouse Management business unit. But whatever the cause, the basic rescue strategy is the same: “Just like medicine, you do triage,” she said. “The most life-threatening problem we try to solve first. And only then, once we’ve stopped the bleeding, we can move on.”
In McCaskill’s comparison, just as a doctor might have to break some ribs through energetic CPR to get a patient’s heart beating again, a failing warehouse might need to recover by “breaking some ribs” in a business sense, such as making management changes or stock write-downs.
Once the business has made some stopgap solutions to “stop the bleeding,” it can proceed to a disciplined recovery, she said. And to reach their final goal, managers can use the classic tools of people, process, and technology to improve what she called the three most important key performance indicators (KPIs): on time in full (OTIF), inventory accuracy, and staff turnover.
CSCMP EDGE attendees gathered Tuesday afternoon for an update and outlook on the truckload (TL) market, which is on the upswing following the longest down cycle in recorded history. Kevin Adamik of RXO (formerly Coyote Logistics), offered an overview of truckload market cycles, highlighting major trends from the recent freight recession and providing an update on where the TL cycle is now.
EDGE 2024, sponsored by the Council of Supply Chain Management Professionals (CSCMP), is taking place this week in Nashville.
Citing data from the Coyote Curve index (which measures year-over-year changes in spot market rates) and other sources, Adamik outlined the dynamics of the TL market. He explained that the last cycle—which lasted from about 2019 to 2024—was longer than the typical three to four-year market cycle, marked by volatile conditions spurred by the Covid-19 pandemic. That cycle is behind us now, he said, adding that the market has reached equilibrium and is headed toward an inflationary environment.
Adamik also told attendees that he expects the new TL cycle to be marked by far less volatility, with a return to more typical conditions. And he offered a slate of supply and demand trends to note as the industry moves into the new cycle.
Supply trends include:
Carrier operating authorities are declining;
Employment in the trucking industry is declining;
Private fleets have expanded, but the expansion has stopped;
Truckload orders are falling.
Demand trends include:
Consumer spending is stable, but is still more service-centric and less goods-intensive;
After a steep decline, imports are on the rise;
Freight volumes have been sluggish but are showing signs of life.
CSCMP EDGE runs through Wednesday, October 2, at Nashville’s Gaylord Opryland Hotel & Resort.
The relationship between shippers and third-party logistics services providers (3PLs) is at the core of successful supply chain management—so getting that relationship right is vital. A panel of industry experts from both sides of the aisle weighed in on what it takes to create strong 3PL/shipper partnerships on day two of the CSCMP EDGE conference, being held this week in Nashville.
Trust, empathy, and transparency ranked high on the list of key elements required for success in all aspects of the partnership, but there are some specifics for each step of the journey. The panel recommended a handful of actions that should take place early on, including:
Establish relationships.
For 3PLs, understand and get to the heart of the shipper’s data.
Also for 3PLs: Understand the shipper’s reason for outsourcing to a 3PL, along with the shipper’s ultimate goals.
Understand company cultures and be sure they align.
Nurture long-term relationships with good communication.
For shippers, be transparent so that the 3PL fully understands your business.
And there are also some “non-negotiables” when it comes to managing the relationship:
3PLs must demonstrate their commitment to engaging with the shipper’s personnel.
3PLs must also demonstrate their commitment to process discipline, continuous improvement, and innovation.
Shippers should ensure that they understand the 3PL’s demonstrated implementation capabilities—ask to visit established clients.
Trust—which takes longer to establish than both sides may expect.
EDGE 2024 is sponsored by the Council of Supply Chain Management Professionals (CSCMP) and runs through Wednesday, October 2, at the Gaylord Opryland Resort & Convention Center in Nashville.