Last year, the company generated almost $3 billion in revenue worldwide. Daifuku has approximately 6,500 employees globally, while Wynright has around 500.
Material handling is the aspect of logistics which is mainly concerned with the movement of boxes from one place to another, usually over short distances, such as from a truck to a warehouse, and then from the warehouse to the trucks for the dispatching of goods to customers’ locations.
Technically speaking, moving boxes around sounds simple if a little tedious and tiring – if you’re human, but when you’re moving billions of boxes a year around the world, as Daifuku Wynright does, and they’re boxes of varying shapes and sizes and are required to be delivered within ever-decreasing timescales, material handling can be an incredibly complex logistical challenge – especially if you’re a robot.
And that challenge is being progressively resolved with increasingly sophisticated solutions that feature ever-higher levels of automation and computing.
In this wide-ranging and exclusive interview, Robotics and Automation News speaks to Giovanni Stone, director at Wynright Robotics, and Joe O’Connor, director of marketing at the company.
The shortest distance from A to Z
In the past, a lot of material would be handled by human workers, but they are becoming very difficult to recruit, and this is one of the factors driving automation in the industry.
“You might assume that people are moving towards automation simply to go faster, and that is certainly one of the reasons,” says Stone.
“But what we find is that robots are generally about as fast or marginally faster than humans at some tasks, although there are some tasks in which robots far exceed the rates that humans can achieve.
“What we find, especially here in the US, is that recruiting the right people to do these jobs in warehouses is very difficult.
“We have clients that have had requisitions for employees for a year or two years without being able to find the people to fill the roles, and a lot of them have no other option but to say, ‘Okay, how else can I do this job?’
“I think this is a trend that is going to continue, and actually continue to get worse, especially for the logistics industry. I think this is driving automation as much as anything else.”
Pallet of many boxes
Increasing levels of automation in the material handling sector has led to the installation of industrial robots for loading and unloading trucks, autonomous mobile platforms and vehicles for transporting, storing and retrieving goods, which are usually in boxes, which, in turn, are usually on pallets.
Pallets – those wooden platforms, as pictured above – are almost all of an internationally agreed standard size, 48 inches by 40 inches.
Other internationally agreed standards cover shipping containers – those huge metal boxes which can be seen in their hundreds and thousands at ports around the world, and individually on the backs of trucks.
Shipping containers are mostly 20 feet or 40 feet in length, as stipulated by the International Standards Organisation.
There are a few other ISO standards for shipping container sizes, but when it comes to the actual boxes on the pallets inside the containers, they can be of a bewildering variety of sizes and shapes.
This makes unloading the pallets in a typical warehouse a complex job that – at the moment – only a human can do affordably enough and quickly enough.
However, with advances in vision technologies and newer grippers or end-effectors for robots, perhaps pallet unloading is another job that the machines can take over from humans.
These sorts of technical issues are currently on the minds of Stone and O’Connor, as well as probably many other people in the material handling industry.
Stone says: “We’ve recently released a product called ‘Mixed-Case De-pal’, or automated mixed case de-palletising, that we’re providing to some of the parcels companies – like UPS and FedEx – and that product is designed to take a pallet of many mixed cases regardless of size and shape and it will unload that pallet onto a conveyor system for sortation.”
Some industry observers are suggesting that there should be internationally agreed rigid standards for box and packaging sizes. But others – especially product designers – tend to favour more unusual or even unique packaging and boxes, which might make things more complicated for material handling companies.
But whatever the solution is, the current situation does seem to lead to a significant amount of waste – both of space and material.
The difficulty is that there are so many varieties of products, available through countless e-commerce outlets, that people order things in many different quantities and sizes.
Stone says: “A lot of our customers have told us that the biggest complaint that they get is when a customer receives that one very small item in a very large box – it just seems so unbelievably wasteful. It’s one of the frustrating things that any consumer will tell you about. But I think there are a lot of companies out there that are working to address that problem.
“In the past, de-palletising was a very complex manual job – something a human worker had to do. But we think at that developing non-specific type tooling that is flexible is more important than forcing the products to be a certain size or shape.
“When you get into the manufacturing sector, that’s where you get the gains for standardising and containerising.
“You’ll see that a lot in automotive, where they’ll take parts and put them in standard containers. That’s something that works well when you’re making the same thing every day, and when you’re doing the same thing over and over.
“I’m just not sure that in logistics that we’ll ever get there.”
Shuttle’s successful launch
Wynright describes itself as an integrator of material handling solutions – the company earned official robotics integrator certification from industry body Robotic Industries Association about five years ago.
However, many of Daifuku Wynright’s solutions feature technologies that the company designs and manufactures within its operation.
For example, through its many business units, the company manufactures hardware such as conveyors, rails, fixed and mobile platforms, and automated guided vehicles, to name a few items.
It also builds something it calls the Shuttle Rack (shown as a computer-generated image in the main picture), an automated storage and retrieval system which the company claims is the best-selling ASRS in the world.
And the company has a whole division dedicated to software development.
Stone, who manages the Wynright robotics business from concept through to implementation from an engineering execution perspective, explains. “We really look at ourselves as robotics integrators,” says Stone.
“From an implementation and business perspective, it makes sense for the space that we’re in – which is primarily distribution or logistics – to basically take existing technology and reintegrate that with our own algorithms and 3D vision software.
“We certainly have custom hardware capabilities for tooling and designing the overall systems, but we don’t do anything with robotic arms – we always buy those because there are many people doing it. It would be a fool’s errand to try and figure out how to do that on your own.
“One of our defining strategic advantages is that we have a team of engineers that help us to develop our own 3D perception technology.
“That is what enables us to do things like robotic truck loading and unloading, and where the robot is moving in a non-deterministic way.”
By “non-deterministic”, Stone means that the robot can do different things on different runs, depending on what the robot itself decides.
Of course, the robot itself needs to be programmed accordingly, and 3D perception is the critical technology which enables these non-deterministic capabilities.
That is, the robot has to initially “see” what it is dealing with before it makes a determination, based on its programming.
Stone seems particularly interested in 3D perception technology, which may open up even more applications for robots within the material handling sector.
In fact, advances in programming or software, in general, is increasingly becoming the route to productivity gains and efficiencies – the hardware is mostly ready for most applications now, it’s just a matter of how to instruct it using computer code.
O’Connor adds: “Daifuku Wynright is an integrator, but we’re also a manufacturer, which makes us unique in the market that we’re in. We manufacture many of the products we integrate into our systems.
“For instance, the platform the robotic arm is placed on for loading and unloading of trucks, and we design and develop the mobile carts.
“The platform has the 3D perception integrated into it, and so does the mobile cart, which is what drives into the [shipping] container to unload it.
“And then, of course, there is the conveyor that gets integrated into a larger system.
“And at the other end of the robot, we design and develop the end-effector, the type of hardware system component that can pick up and move material.
“Also the robot transport units for linear palletising, where the robotic arm sits on a platform which moves back and forth along a rail or track system.”
Stone says 3D perception system that Wynright has developed works at a much longer focal length – at around 5 to 10 metres – than systems used by other companies in the market, which tend to concentrate on close inspection, next to or immediately overlooking the conveyor, which is probably less than a metre.
The Amazon effect
Arguably, one of the reasons for Amazon’s amazing success is its decision to use vast numbers of robots in its warehouses, automating its operations to a point where it is able to offer super-fast delivery times.
And given that Amazon is said to have received 40 per cent of all online shopping orders in the US over the Christmas period, it’s no wonder that many companies are looking at the e-commerce giant to learn from its innovations.
The phrase “the Amazon effect” has come to mean massive ongoing disruption in the retail market – both online and on the streets.
Stone says Amazon’s system enables the company to be “extremely efficient” and capture large portions of the market, and its operation has caught many others “flat-footed”.
“People’s expectations are really what is changing about e-commerce,” says Stone. “Consumers are expecting their orders to be delivered the second day, next day and sometimes same day, depending on the products that are being delivered.
“A lot of companies that maybe didn’t have that infrastructure have been caught a little flat-footed and now they have to change. Or they have to go to Amazon and ask them to distribute their products for them.
“I think that what we see is that now the problem with e-commerce distribution is purely volume. How do you get that much stuff out the door and to people?
“That then leads to further supply chain issues for the delivery companies and for the parcel companies – how to handle this mass influx of products and increase in volume.
“It used to be that you ordered something from a Sears catalogue or something and six weeks later it would show up and that was just how it went – we were really good at waiting.
“But as consumers, we’re no longer good at waiting, and that leads us to that volume problem in the industry.
“I would say that now those problems are being solved. I don’t think anyone has the magic solution yet, but I do think there are a lot of people with innovative solutions.
“I think what’s important to remember is that there is no one-size-fits-all solution to any of these problems.
“There is the right technology for any problem that’s out there, but there is no one that solves it for everybody.
“I always think it’s interesting that when I speak to the autonomous guided vehicles guys and I hear, ‘A future without conveyor’.
“But I think there will always be applications where a conveyor is well-suited, and there will be other applications where an AGV would be more well-suited.
“It all depends on rates and the amount of effort it takes to interact with the technology.
“The AGV guys, they have a very new technology that will continue to develop over time, and the conveyor’s been around since 1905 or something like that, so it’s had a lot of time to develop and find solutions to these different markets.”
Stone credits the automotive industry – which buys around half of all industrial robots in the US – for the increased availability of robotics technology.
“Because of the almost insatiable appetite for robotics in industries like automotive, the cost of robot components and robot arms have been dramatically reduced, making them more attainable, helping people get to the return on investment they are looking for.
“Additionally, I would say that one of the technologies and one of the changes that have been made in the past three to five years is the ability to use what’s called collaborative safety in a robotic environment.
“It used to be that when you put a robot in to any environment, it would have to have all sorts of sensing and cages and all sorts of things to protect users from the robot.
“Some of those regulations have been relaxed a little bit because of advancements in 3D scanning technology such as area scanners and scanners around the robots, so instead of setting up a fence, I can set up a scanner that will allow somebody to walk close to the robot and the robot automatically knows to stop because that person is close.
“I think some of the other advancements have been more software-driven. So as people get into integrating robots into larger systems or the real process that is driving this – the warehouse control system, or what we call Wynsoft – is important because that robot has to take data and information from the host system and then process that data and translate it into some sort of action that allows that robots to provide value.”
Wynright offers a range of software for managing warehouse operations, and they can either be used as standalone systems or be integrated into customers’ existing warehouse management systems or whatever software they have.
O’Connor says that a lot of the company’s business is now with companies which are looking to upgrade their warehouses, perhaps to increase levels of automation and efficiency.
“The big markets we serve include retail and distribution, footwear distribution, food, beverage and grocery distribution, direct fulfilment, general distribution, big-box retailers, and so on.
“In addition, Wynright manufactures the Automotion brand of conveyor that customers can use with our robotic packages, but that’s also the conveyor that we integrate into all of our systems.
“We manufacture mezzanines and other structural support pieces that we integrate into our systems.
“The robotics technology is integrated into our systems.
“And we also have a voice-directed and pick-to-light that we integrate into the different products and services we offer.
“And then all of the little divisions of services, from anything such as design engineering to application engineering, controls engineering, software engineering, and so on.
“We’re all-inclusive. We take the project from concept to execution for the entire system.
“When I say execution, there’s the commissioning and installation and everything in there.
“But also we’re also often brought in as an adviser to provide some analysis.
“For instance, there’s asset modernisation, where we’re called in as the experts and we help them – not necessarily in implementing a physical solution but oftentimes it’s the process, where we are able to share our experience with them to help with processes.
“Sometimes we just do software upgrades, things of that nature, wholesale support.
“Asset modernisation is another big area for us. When times are tight, customers don’t want to necessarily replace an entire system at the cost of millions. We can show them how to extend the life of some of their ageing assets, whether it be with software upgrades or things of that nature.
“If you think of Microsoft, where they continually upgrade their platform, it’s a similar type of thing we are able to come in and do – except we can upgrade the hardware and the software.”
Stone adds that one of the key areas of innovation and upgrading is going to be in end-effector technology, which will see the emergence of more sophisticated, multipurpose grippers.
Stone says: “I think there are some very interesting advances in gripping technology that will drive the sector in the future and help us to continue going forward, and you only have to look at companies like Soft Robotics to find those types of technologies.”
At the moment, he says, vacuum and suction technology is used in many applications and that has been around for quite some time.
Virtual material handling
One of the other technologies both Stone and O’Connor are enthusiastic about is virtual reality, when it is used in the design and development of material handling systems and warehouse facilities.
Stone provides an example of a project which involved simulation. “It was both a greenfield [undeveloped] and brownfield [partially-developed] site from a reconfiguration perspective that we needed to simulate and understand how that process was going to work.
“I think the introduction of simulation usually happens once you get to a point you feel like you have a design that’s going to work well. It can be a time-consuming process.
“So, once we whittle down to maybe one or two options, we’ll then simulate those options to provide details about how things are going to work, and we may simulate the entire thing or we may just simulate portions of it in order to make sure we’ve validated the different aspects of the design.
“If it’s a big customer with a lot of risk involved in their process in general, they actually emulate the system. So that’s one that’s a little more involved, where we’ll take their WMS, or WES, or WCS, and connect it to our controls – and we’ll have a PLC set up to provide some feedback to the system – and we may run the system inside of the simulation with all of the host systems included.
“That’s the best way to mitigate risk in a system in general and it helps us to avoid inefficiencies and so on.
“But that process is not always perfect, of course, because there are things you learn when you are out in the field, but it does help us to implement things a little faster.
“In other parts of Wynight, such as baggage handling systems, they do a lot of this for the baggage handling industry where they basically get, like, two hours to implement a new system instead of a matter of weeks and the whole thing has to be simulated and emulated to the final solution before they are able to actually implement something in the field.”
O’Connor chimes in, adding: “I’ve actually played with some of the simulation and emulation applications that they have and they’re impressive.
“You could put on one of the virtual reality headsets and you are in the system. You’re not just testing rates, you’re an actual handler in the system where you reach out with a sensor and pick up a bag or move a bag from one spot to another and it’s really quite impressive.”
O’Connor also provides an example: “We had a customer that was great at what they do, but they needed help figuring out the best way to do it.
“And there’s usually a couple of things to consider – there’s always cost, of course, and another big consideration is the footprint: how much room do they have in their existing space, or are they building greenfield and the footprint doesn’t matter?
“In this particular case, it was not greenfield, it was brownfield, and we were constricted on space and things of that nature, but what it came down to was our design team came up with a couple of different ways of solving the problem.
“And it goes back to what Gino was talking about earlier – it’s not always the conveyor that is the cure. In this case, there were two solutions.
“The problem at hand was they needed to load their trucks in a specific order so that it became easier for their customers to manage the shipment.
“They were always the best price, best quality, but the biggest headache to their customers was that when a truck came in, it took several men several hours to get this thing taken care of.
“And what we gave them was two different scenarios: one was using a pre-sort process, which was literally miles of conveyor where product would first get sorted by groups of 100 and then broken down again and again until ultimately the bar code was able to be sorted in a one-to-10 position. This would allow us to reverse load onto a truck. And that would all be done with conveyor.
“But there’s another piece of equipment manufactured by our parent company, Daifuku, in Japan, the loader and the mini-loader ASRS, and the Shuttle Rack M.
“The Shuttle is a specialised, mini-load automated storage and retrieval system, or ASRS, and what the Shuttle does is that it takes a product and assigns it a space in a vertical rack.
“Instead of a typical ASRS, where you have a crane that goes down and picks location to location, a Shuttle has a vertical lift in the front, and the product gets loaded into the vertical lift, elevated to whatever tier it has to go to, and then gets picked up by an individual Shuttle at that layer and brought down to the position and inserted in.
“So it’s still that XY axis that a single crane would access, but in this case, we have multiple Shuttles so the system is much quicker – it can put away and retrieve product much faster.
“And what it came down to was that the Shuttle has a much smaller footprint, it was very quick, product could be retrieved in whatever order they needed and then loaded on the truck, so the difference came down to cost and footprint.
“We also had to show the customer that both of these systems would work. Then we showed them with the emulation process how it would work.
“That’s an example of the practical use of the simulation technology.”
The most important thing about a solution, whether it’s simulated and emulated in advance or not, says Stone, is that it “must be effective 99.9 per cent of the time – it can’t be effective 97 per cent of the time, it can’t be effective 96 per cent of the time, it’s got to be at least 99.9 per cent”.
Stone says: “Wynright Robotics has been around for about 20 years, and we have a lot of experience working with clients who are trying to apply automation to a problem and get it to the point where it is truly autonomous, and where it works that 99.9 per cent of the time.”