NextGen Supply Chain: The Robots are Here

This is an excerpt of the original article. It was written for the January-February 2018 edition of Supply Chain Management Review. The full article is available to current subscribers.

January-February 2018

If you frequent supply chain conferences, as I do, you’ve probably noticed that some of the best-attended sessions are the ones that focus on emerging technologies—or what we’re calling the NextGen Supply Chain. You may have noticed something else: While topics like Big Data, artificial intelligence, augmented reality, blockchain and robotics play to standing room audiences, there’s a lot of confusion about what to do with the information. At the 2016 APICS conference, one member of the audience asked a direct question at the end of an excellent session on Big Data by Hannah Kain, the CEO of Alom: “This sounds great. But there’s not a…

Browse this issue archive.

Already a subscriber? Access full edition now.

Need Help?

Contact customer service

847-559-7581   More options

Not a subscriber? Start your magazine subscription.

Robots have been a fixture of science fiction books and movies for decades. Perhaps our infatuation with them stems from some innate need to reproduce humanity in the form of a machine. Or, perhaps, it comes from a desire to create machines that can take over the mundane tasks of our everyday lives. Whatever the reason, the fact is, the idea of robots has prompted innovators to go out and build the robots that we see today and will continue to see into the future.

But robots aren't just a plot device in sci-fi or playthings for the nerds in the “Big Bang Theory” when they aren't in a comic book store. In the business world, robots have been in use in industrial manufacturing since their introduction in 1962 in a GM automotive plant. But, until recently, they haven't gotten much traction beyond the assembly line or one-off use cases. In part, that's because the industrial robots used in manufacturing differ quite significantly from the new generation of robots making their way into a broader range of industries and applications. Industrial manufacturing robots are designed to operate at a high rate of speed, precision and strength. While these traits are good for automating manufacturing processes, they don't translate well to other areas of the business that require more human interaction. Indeed, they are part of the reason why industrial manufacturing robots have to operate in “cages:” areas where human workers aren't permitted to enter during operation.

It's safe to say, however, that the development and maturity of industrial robotics has helped to pave the way for the new generation of robots that is quickly being introduced across business functions and industries. Robotic technology has been on a continuous path of improvement over time as robotics manufacturers look to improve their technology and layer modern digital technologies onto the mechanical elements of robots.

Indeed, over the past several years, robots have increased their level of intelligence and flexibility as machine learning and artificial intelligence (AI) have been built into the systems; there have been significant improvements to the mobility of robots, while the ability to connect robots via IoT and enhanced safety measures have also been built into robotic devices. The result of these innovations is a new style of robot that is more suited to work side by side with humans rather than be segregated in a human free zone.

Referred to as collaborative robots, or co-bots, the new generation of robots are safe, smart, collaborative and can work side-by-side with people. They have become immensely sophisticated, resulting in modern robots that are increasingly being used as a mechanism to automate business processes that have historically been incapable of automation. This next generation of co-bots is one of the factors driving robotic adoption in new industries and across new points of use.

Disruptive processes

One business area ripe for business process disruption enabled by robotics is supply chain execution, especially in order fulfillment processes in the warehouse. These processes typically involve a high degree of human involvement as well as a tremendous amount of movement throughout a facility. Now, it's not as if robotics have been absent from these areas in the past; there are use case examples, but none at a large scale across supply execution. Those organizations that have introduced robots into their warehousing and fulfillment operations have delivered added value including productivity improvements, efficiency gains, the capability to better scale up/down with demand spikes and the ability to improve customer service levels.

The most familiar example of robotics in the fulfillment process is at Amazon. The e-commerce giant acquired Kiva Systems (now known as Amazon Robotics) in 2012 for $775 million. Since then, Amazon has continuously expanded their use to upwards of 80,000 robots across 25 distribution centers. Through their deployment, Amazon has been able to accelerate delivery times and reduce fulfillment related costs. According to a note published by Deutsche Bank, the deployment of the robots equates to a roughly $22 million per year savings in facilities where they are in use, or an estimated 20% reduction to operating costs.

If Deutsche Bank's estimates are close, Amazon has proven that there is tremendous value to be gained through the use of robotics within the fulfillment center. However, the acquisition and subsequent privatization of Kiva Systems created a void in the market. Companies that were interested in this type of robot for their fulfillment processes were no longer able to procure Kiva Systems robots; those who were already using Kiva were confronted with the eventual loss of support for their investment. That was a real disincentive to anyone considering an investment in robotics from another start-up that might be similarly acquired and taken off the market. On the positive side, business, like nature, abhors a vacuum. Kiva's exit from the market created an opening for opportunistic companies to create innovative new robotic technologies and solutions to fill the void, and to improve upon the technology itself.

Some of those companies are offering Kiva-like solutions, but we are also seeing the emergence of new competitors taking a decidedly different approach to robots in the fulfillment process from Kiva. The result is that fulfillment operations have a variety of robotic solutions from which to choose. And, as with most technologies, the first step is for organizations to define their needs in order to choose an appropriate technology to meet those demands.

For some, a Kiva-type model will work quite well. These utilize fast-moving robots that shuttle entire racks of inventory from a segregated section of the fulfillment center to a picking station, where a picker selects the inventory needed to fill an order. After a pick, a robot returns the rack to back to the floor and moves on to the next pick. Think of this as a goods-to-person, or better yet, a rack-to-person, solution. Solution providers offering this model include GreyOrange and Swisslog's CarryPick mobile system. A rack-to-person model is best suited to high throughput facilities where speed is the most important element. The benefits include the ability to rapidly move product to picking locations and accelerate fulfillment cycles. However, the rack-to-person model also has its drawbacks. For example, it requires some facility modification to create a segregated area where the robots can safely operate and it requires a guidance mechanism to ensure that the robots operate within the appropriate spaces. These systems are not necessarily collaborative because humans aren't allowed to work in the same aisles where the robots are operating. One final drawback is that with these models, half of the movement is spent returning racks after a pick, essentially retaining 50% of the wasted movement in the process.

That is one model. Still other companies have emerged with collaborative robots designed to complement traditional picking processes. Some notable vendors in this space include Locus Robotics, Fetch Robotics and 6 River Systems, to name a few. The devices from these companies have been designed to operate on the floor in the same spaces where associates are at work, often in collaboration with those associates. In this model, the robot is assigned a task by the WMS and automatically navigates to a pick location. When it arrives at a location, an associate will pull product from a shelf or carton and place it in a bin or tote on the robot. The robot is then directed by the WMS to the next pick location until all the items for an order are in the tote and ready for delivery to the pack station. Once the pick is complete, the robot moves onto the next pick order rather than returning to drop a rack.

This model can be very effective for fulfillment centers with a high variety of orders, including a significant number of multi-product orders. Because the robot is responsible for all of the non-value-added travel between tasks, associates on the floor are more productive because they remain in their zone where they are focused on picking items from the shelves. What's more, they can typically operate within an existing infrastructure, which can help minimize the upfront investment and accelerate deployments and go-live. The drawback: The robots sacrifice speed for safety, operating at slower speeds than their non-collaborative counterparts. That's not necessarily a bad thing, but it's certainly a point worth mentioning.

Sophisticated navigation systems are also essential in the collaborative mobile robot space. That is because co-bots are designed to autonomously navigate their way through a fulfillment center and around potential obstructions. Once the facility is mapped within the robots guidance system, the robots are free to roam where other systems require some guidance support, such as a beacon, white line on the floor, or some other type of physical marker to help guide the robots. However, co-bots must also be capable of decision making when it comes to selecting an initial route or creating a new route when the robot encounters an obstacle, much the way a GPS system recalculates to a change in the original route.

As previously mentioned, both approaches are an effective means to deploying robots. The question becomes which approach is most appropriate for your individual scenario. Clearly, not all fulfillment center operations are set up the same. Just as its common to find more than one kind of picking process and technology in a conventional warehouse, different facilities may identify different strategies to achieve the increased value. For example, a mix of high-speed rack-to-person robots and co-bots may be the way to go. The objective of introducing robots into the fulfillment process is to improve the overall operation by reducing costs, improving productivity, improving efficiency or some combination of the three.

Why now?

Regardless of our fascination with robots, no organization is going to deploy technology for the novelty of it. There must always be a business case before a company makes a decision to invest in new technology. Robots are no different. The business case for their deployment is being driven by the value proposition of reducing costs, increasing productivity and improving efficiency. Sub-elements that contribute to achieving value include the ability to increase speed and inject flexibility into the processes, which support the objectives of improving productivity and efficiency.
Still, the larger question remains, why now? Robots have been around for decades, so what is the breakthrough that is finally enabling robots to deliver value across the fulfillment center?

There are several ways to address this question.The first is to look at how the advances in related technologies have led to improvements in how robots perform.

They include:

• Artificial Intelligence (AI) and cognitive computing. AI is often associated with a humanoid robot that looks and sounds like a person. When thinking about advancements in robotics for supply chain applications, the AI and cognitive elements are more related to mobility. The collaborative mobile robots that have made their way into modern fulfillment centers are able to autonomously navigate a facility because they have the onboard intelligence to identify an obstruction and make a decision about how to respond to that disruption. Such an advance in mobility is quite significant in enabling mobile collaborative robots to add value to the fulfillment process.
• Internet of Things (IoT). IoT provides a mechanism to communicate with and capture data about an operation in real time. Robots are a connected asset and as such create a tremendous amount of data about the business processes they are involved in. Through such data capture, organizations can now capture data about previously manual business processes. But, it's more than just data capture; through IoT, we have the capacity for a WMS system to deliver a pick list directly to a robot that then knows exactly where it needs to go as well as the most efficient way to get there. Finally, consider the inventory management capability: Through connected robots, fulfillment centers have the opportunity to leverage connected robots to send inventory signals directly to the WMS.
• Cloud. The Cloud is helping to drive improvements in robotics. Through the Cloud, coupled with IoT, organizations are now able to monitor, manage and even operate robots from anywhere they have a connection to the internet. This point is drastically enhancing the usability of robots in the workplace and has even helped robot manufacturers to enhance their service to their customers. For example, through the Cloud, manufacturers can remotely monitor the health of their clients' robots. They also have the ability take control of a mobile robot that has encountered a situation that it can't resolve on its own and navigate a robot back towards appropriate operations without inconveniencing the client.

The second “why now” point is the continuous innovation in the field of robotics. Robot vendors have been able to build upon the technology over time to develop modern robots that significantly outperform the robots of the past. A big part of this continuous innovation is the application of related technologies (as mentioned above) to the field of robotics. Additionally, we must consider how robotic vendors have continued to improve the mechanical capabilities of robots. Consider the recent video from Boston Dynamics that shows an Atlas robot performing a series of box jumps, ending with a back flip. In fact, if you haven't seen this look it up, it's pretty cool. * While this motion does not perform a function that is useful in the supply chain, the display of dexterity showcases the rapidly evolving physical capabilities of robots, with each new innovation being built upon the existing set of robotic capabilities.

One final answer to the “why now” question is the readiness of the market. This may be the most important point in this discussion. As Daniel Theobold, co-founder and chief innovation officer at Vecna Robotics, recently mentioned: “we have been building and improving our robots for more than 20 years, with the military and in medical facilities, so that when the broader market was finally ready we would be ready to meet the markets needs with a mature suite of products, the market is finally ready for this technology.” Companies today recognize the value of robots in the supply chain, and are increasingly looking to deploy the technology to improve their ability to serve their clients and build a competitive position in the market.

Ultimately, the “why now” is not answered by any one of the afore mentioned points independently, but rather through the combination of technology interplay, robotic technology maturity and market readiness. When you bring together a market that is ready to accept robots, robotic technology that is mature and continuing to innovate, and an abundance of related technology that are helping to advance robots, you arrive at the point where robots in the supply chain will become required to compete rather than a means to enable competitive advantage.

This already happening. Since its acquisition of Kiva Systems, Amazon has leveraged robotics to create a competitive advantage, and its competitors are now deploying robots just to keep up in the fulfillment market. “The ever-increasing demands for faster and more accurate order fulfillment requires IT tools that accurately track and measure both human and robotic performance,” notes Bruce Welty, chairman of Locus Robotics. Companies that are better able to extract value out of the data related to their fulfilment processes will be in a better position to drive value for their customers. Robots are providing a mechanism to capture such data to drive this value.

The future of robots in the supply chain

Based on our research at IDC, we believe the future of robots in the supply chain is bright. A growing market demand will continue to encourage robotic vendors to innovate and bring advanced functionality to the supply chain. While robots have historically been built and deployed to perform a specific process or function, the constant state of innovation in robotic development will evolve here.

IDC's 2018 Robotics FutureScape report predicts “by 2019, 25% of mobile robotic deployments will include the ability to add on modular components enabling multiple uses on the same mobile platform, thus delivering up to 30% productivity and efficiency gains.” Through this prediction, IDC is expressing the belief that mobile robots will be able to be outfitted with components that enable multiple business processes to be performed while leveraging a common management platform.

Melonee Wise, CEO of Fetch Robotics, agrees with this belief stating: “We are banking on the future success of the collaborative robotics market as a combination of common hardware platform and custom, task- or industry-specific software. Anyone can build a robot that does one thing well; the real opportunity lies in deploying a modular platform that can be extended to a wide variety of solutions.” As with any technology, innovation in robotics and the use of robotics is paramount. A competitor that embraces change will quickly disrupt the company that sits idle and believes it's at the top.
Speaking on supply chain innovation and robotics, Adrian Kumar, DHL's vice president of solutions design, says: “We believe it’s critical to identify and implement these types of advanced technology solutions in the warehouse so we can seamlessly improve our customers’ supply chains.” He adds that the implementation of a robotic pilot program within the life sciences sector at DHL Supply Chain “will inform the potential for broader deployment across different parts of our business. This is a natural evolution of our robotics program.”

This is a testament to the importance or robotics on the future of the supply chain. Robots are not just an interesting technology for science projects. They are quickly becoming a critical technology that is helping supply chains to innovate and deliver exceptional customer service while improving business performance. Those companies that sit on the sidelines, rather than invest now, run the risk of becoming obsolete in a market that is increasingly reliant on technology—and more specifically robotics—as a mechanism to drive value.

Check out the Boston Dynamics back-flipping robot.

John Santagate is the research director for service robots at IDC. He can be reached at jsantagate.idc.com. For more information, visit idc.com.

SC
MR