The broader human machine interface (HMI) represents the man/machine handshake, the joint at which humanity ends and the humanity’s creation begins. But it’s not a static thing. There are slow, almost geologic forces at work that are constantly chipping away at the HMI landscape, while the occasional deluge event, like the mobile touchscreen revolution, creates a momentous leap forward. It’s a constantly moving target, and it’s moving at an interrupted pace.
But within the packaging and processing space, the HMI is in a unique state of limbo, still lurching forward, but held equidistant between the past and the future of the technology. Tip-of-the-spear consumer HMI products from Apple and Android drag industrial HMIs forward, sometimes kicking and screaming. Meanwhile, the requirement for longevity acts as an anchor to progress. Obsolescence is difficult to plan for in HMIs, as the HMI technological trajectory outpaces that of the machines they serve. Internal forces are at work, too, with the physical size of a touchscreen pad being a limiting factor at both large and small extremes.
Considering HMIs can only exist in a sweet spot within such a host of practical, external and internal constraints, you’d think there would be precious little wiggle room for variety among packaging and processing machine HMI styles and functions. But, the opposite is the case. HMIs have been a real wild West, varying from machine builder to machine builder with little to no standardization. And with a sparse labor market, the need to train operators on a new HMI ‘language’ at every interface – several existing within a single line—becomes daunting.
Is there a common ground? Where can OEMs practically standardize, and where do we have to stay flexible? It’s a difficult tightrope to walk, but design and applications engineers and many leading OEMs are already walking it daily.
The Apple Effect and practical expectations
The consumer mind snaps to the smartphone or tablet when thinking of HMI, as that’s what’s in everyone’s pocket. This can be called an Apple Effect, for lack of a more agnostic term. But it’s important to first set realistic expectations as to what can reasonably be done with HMI in the packaging and processing space, to disabuse the expectation of an iPhone 6s interface as your typical palletizer control.
First, there’s a different mindset at work when using an iPhone compared to interfacing with machinery. According to Bob Williams, general manager, Axon, a division of Pro Mach, his customers and operators alike are addressing a practical necessity, not an emotional one. “Quite simply our machines put a label or a tamper evident feature on a customer’s product. Our customers buy our machines not for what they are, but for what they do, i.e. applying a means to communicate their product’s value to consumers, or providing a safe, secure, authentic product. So at Axon, we recognize that as much as we may love our machines, they are a means to an end. It follows then that for Axon, and every OEM, the HMI is one of the most critical components of our machine; for it is through the HMI that the operator sets the production parameters to most efficiently and effectively produce the best looking finished product.”
Not to mention, Apple technologies are produced in the millions, for the millions. Packaging machinery HMIs are not, so there’s no economy of scale to go into improving the HMI experience. And scale is necessary to keep in mind when translating HMIs to our industry.
“We see people sometimes balk at machine prices, saying, ‘A person could buy a BMW for that price, and a BMW looks a lot better and has a lot more horsepower,’” says Greg Berguig, VP, marketing, PAC Machinery, San Rafael, Calif. “As we know, you can’t compare a $300 Apple iPad, made in the millions, to something made for a specific task, in a comparatively tiny quantity, going through several levels of distribution.”
The perception of high tech mobile moved quickly from an executive-only luxury to an everyman’s tool. Apple ubiquity has taught all of us, machine operators included, to expect the best—the best is the baseline. Those entrenched in the Apple or Android interface are underwhelmed at best when looking at a typical packaging machine HMI, but those folks aren’t looking through an appropriate lens.
Price points are also important to keep in mind in the context of machine variety. According to Dr. Bryan Griffen, group engineering manager, Nestlé USA, and Organization for Machine Automation and Control (OMAC) chair, HMI displays should be commensurate to the machine’s price and complexity.
“You have some machines that are small and less sophisticated, and then extremely sophisticated massive machines that cost $500,000. You don’t want to be killing a fruit fly with a bazooka by splurging for a fancy system on a really basic machine,” he says. “That’s why we always want to introduce scalability when we talk about standardization.”
Looking forward while tethered to the past
Unlike to Apple, backward compatibility is a constant issue to industrial HMIs. PAC Machinery’s Berguig mentions recently looking at Android industrial tablets off which the company’s machines could run—compared to most current industry supplier offerings, this tablet would provide a user friendly experience. But the drawbacks are currently too great.
“One major concern is that even though some software allows you to run a PLC integrated to an iPad, what happens when that iPad model is phased out in a year or two? Manufacturers of controls have an advantage in this department; they are backward compatible for 15 to 30 years. Apple isn’t. ”
Speaking of backward compatibility, Ryan Edginton, President/CEO, All-Fill, Inc., Exton, Pa., speaks of it as a necessary evil. “We actually support the old HMIs to a fault. What I mean by that is, oftentimes, we try to make an HMI obsolete because it is expensive and time consuming to support from the manufacturer of the device. Users of our machinery may be best off upgrading at the point of component failure. Controls do go bad, it is inevitable and I believe this is not only a challenge for us, but for everyone in the industry.”
Yet another consideration in defining the practical rationale for a gap between consumer and packaging HMI is a generational gap itself. Though some operators may be digital natives, executives and purchasers more often are not. The Apple Effect is magnetically pulling operator expectations in one direction, but it has little bearing on the current leadership. As generations shift, and digital natives assume leadership positions, Berguig expects a stronger pull towards more standardized and intuitive HMIs.
“The funny thing is how people now relate to cell phones, as a generation, is the same way that the operators relate to the machinery and machinery controls,” Edginton says. “Our grandparents might have a tough time dealing with the latest and greatest cell phone features, but when you talk about HMI, you have to go with what’s simplest for the operator. For example, an operator who worked on a machine with an alphanumeric keypad for years may not want to switch because it is what they are most comfortable with. The younger generation of operators however gravitates and feels connected to the machine that resembles the phone in his pocket. We try to keep our software relatively consistent across our product line so our customers with multiple machines can easily navigate the screen from machine to machine.”
Learn to crawl, learn to walk
In the mean time, OEMs are putting HMIs through their paces, taking industrial design cues from the Apple models, and moving into what Berguig considers a Gen 2.0 of the HMI. The first generation was strictly utilitarian, just getting them to work. Most OEMs have crossed that threshold.
“What we always ask ourselves is, “If there is a redesign, how beneficial is it to end users?” “Does it have value?” Edginton asks. “There are a lot of neat features and programs that can be loaded into an HMI, but is it actually usable for the operator, or just noise?”
The next steps are to get them to work well, efficiently, while borrowing from industrial design best practices.
“The key is going to be intuitiveness for multiple levels of operations,” says Kevin Keller, managing director, K2 Kinetics, York, Pa. “I’m an Apple guy, so naturally I talk about the Apple Effect. If you hand me an Android, I can’t find a home screen, so even in the consumer world, lack of standardization can be a difficulty. The variety in our industry greatly compounds the need for intuitiveness.”
“We want to see it simplified as well,” adds Axon’s Williams. “An easy navigation that is only ever two or three steps away from any other screen. We need to eliminate the maze, so it will only be two navigation buttons get to a diagnostic screen, or three buttons to get to a video to see how to clear the jam. There are a lot of good HMI companies out there but quite frankly, it’s not the iPhone. I look at operators, I think that everybody knows that intuitive screen, that intuitive feel. That’s where we are pushing some of our suppliers to go, and that's what I want the experience be for the operators. An operator familiar with the HMI is going to run our machines more efficiently.”
The automotive world is a good place to look for examples. Every car manufacturer began with digital HMI display differently, and they all started off pretty clumsily. After a few generations (and we’re not there yet), the useless and awkward features drop away, while the useful are retained and refined. This is a Darwinian force acting on all auto HMIs, though it’s at a glacial pace. Features co-evolve among different brands, as they are exposed to the same external consumer pressures. So, end users themselves become an external standardizing force.
That said, while the variety of OEM HMIs constituting Berguig’s “Gen 2.0” all work, may be somewhat intuitive, may even be quite slick, there are still no two that look alike. To continue the car HMI analogy, only true road warrior traveling salespeople will switch rental cars frequently enough to notice the carousel of auto HMIs. The rest of us buy our car, learn our HMI, and worry about relearning in three years’ time or more. In an industrial packaging or processing setting, operators are moving from HMI to HMI in a packaging or processing line almost daily. Nestlé’s Griffen breaks down this common scenario from an end-user perspective.
“We imagine an operator getting sick, so we bring in a new person from another similar line. We chose this person because he knows the machine and we don’t want our efficiency to go down. The problem is, that person has version two of the machine, and they changed the menus, they changed the HMIs. It takes him or her half a day to figure out where everything is on the new machine. It is supposed to be the same as the machine he had on his line but it's not. It never is,” says Griffen.
Standardizing out of self-interest
And standardization isn’t just an end-user issue. Craig Souser, president/CEO, JLS Automation, York, Pa., who has gravitated to PackML as a standardizing force, points to good, old-fashioned self-interest and self-preservation as another invisible hand at work in the evolution of packaging and processing machinery HMIs.
He initially implemented PackML across the board at JLS because his machines have multiple HMI suppliers, and each machine winds up with a different HMI implementation from programmer to programmer, not just from application to application.
“So our service guys had a need for standardization. Even if an engineer left, somebody else had to pick up where the previous left off and try to modify whenever there is an issue,” Souser says. “So initially, for our own benefit, we needed to have a standard. And if one already existed, why should we invent our own? So, we adopted PackML. We started with it for our own sanity.”
While internal forces drove Souser toward PackML, it has become a feature that they try to communicate with their customers.
“We are now trying to make sure we broadcast it to the customer, and let them understand that it's a major benefit to them. But, at least in part, our adoption was selfishly motivated,” Souser says.
K2 Kinetics’ Keller agrees. “We see the need for synergy from both sides. We see the advantages and the headaches that the OEMs and end-users are dealing with,” he says. “I think someone, at a higher level in any organization needs to come out and say ‘this is the best practice.’ Just as now companies are adopting Industry 4.0, there will be a critical mass reached where people will agree on best practice to drive standardization.”
HMI 3.0?
“Who will assume that task of standardization?” Berguig asks. “Will it be us, the OEMs? Will it be controls manufacturers? Or will the end user pull us that way?” For Keller, the change agent will be OMAC (which created PackML), PMMI, or someone like it. But he says that OEMs and their suppliers have their own roles to play in this game.
“It’s going to be a big task to standardize, and it’ll be a multi-step process. There are so many SCADA platforms out there that cost from zero to thousands of dollars, and people are buying everything. Until those guys start pushing simpler-to-use set-up features that follow a standard protocol, it’s hard for any OEM to make a move. But we can make small steps along the way,” says Keller.
He references OEMs that are currently standardizing modes in simple ways, as simple as a background color needing to be uniform in a run mode, or emergency state that should always look a certain way. But it’s still herding cats in getting multiple OEMs, serving multiple markets with multiple functions at multiple price points to agree to a standard.
While OEMs are moving HMIs along to Gen 2.0 in their own ways, a smattering of evidence at PACK EXPO Las Vegas 2015 indicates a unifying standard might be gaining traction: OMAC’s PackML HMI spec. This could be seen as the third major deluge of the slow HMI landscape evolution, or HMI 3.0.
OMAC, PackML, and scalable standardization
Nestlé’s Griffen, partnering with other large CPGs, worked together to create a specification, using PackML, that could stand to be the keel beneath the next generation of industrial HMI. Most importantly, the spec is scalable.
The fully PackML-compliant document is part of the tool set that’s available through OMAC for building PackML enabled machines [published at omac.org]. It defines the navigation screen, and what it should look like. It shows a top bar, a body bar and bottom bar, and it tells the OEMs’ HMI screen designers what should be in each of those bars in terms of navigation buttons or date and time, what kind of formats it should include, what the buttons should be.
“The key thing is we define what the different buttons should be and what order they are in so that no matter what machine the operator goes to, he or she knows the third button over on the bottom bar will be the alarm button,” Griffen says. The top bar features the company name, date, time stamp, machine information, and what mode of operation it is in. The navigation buttons on the bottom are always the same as well, including an alarm band. The centerpiece is where flexibility remains. The machine builder has free reign to use that space as is needed.
“Every machine builder has their own graphics and styles, and that’s OK. We understand that, we are not trying to be iron-fisted about the whole thing,” Griffen says. “It is entirely up to them what goes in that centerpiece because it is their machine, but now, the operator just inherently knows how to get into setup. He knows how to get to the machine parameters. He knows where to go to make this machine operate. He knows what level he needs. He knows if he needs to change a certain setting, then he’s got to get on the phone and he’s got to call his maintenance guy to get in there. Or he has got to call his supervisor to come in and change because he doesn't have the rights. But he knows that before he even has to go in to do it. He can already call and say I am going to be changing this because of what’s going on; can you please send a maintenance guy to come make the right adjustment before he even gets into it. So, we can do better planning.”
As far as adoption, a few HMIs touting PackML compliance were already visible at PACK EXPO, including on Axon’s Lanzara labeler.
Nestlé and other CPGs may be forcing the issue, as well. According to Griffen, Nestlé is starting to specify PackML-compliant HMI in all of its new equipment. As the major CPGs and food and beverage packagers line up, Griffen predicts an eventual tipping point of adoption that will make PackML HMIs the de facto standard.
“One piece of feedback that we have gotten from several of the big OEMs is that they have always wanted the end user to tell us what they want the HMI to look like, but they never do. So, OEMs take a guess, which is frequently wrong,” Griffen says. “An OEM will keep selling until the machine is on-site and the end user will want to add this, we want to add that, move one control to another spot on the HMI. So, having a spec up-front should make their life easier because they know they can comply to what we need to run that machine.”
Asking more of HMI output
Now that touchscreen, flat panel HMIs are almost ubiquitous for machines of a certain sophistication level, what that screen is able to display has become a contentious issue. Troubleshooting video functions, or video-rich instruction manuals, are often championed as potential manna from heaven for both the OEM’s service departments, and the end user’s uptime.
“I think that’s the future,” says All-Fill’s Edginton. “I think that eventually there is going to be a help or troubleshooting section loaded in your HMI. As our business has increased over the years there has never been such a demand on our service and post-sale support team. A printed version of the user manuals has become virtually useless. Everyone is looking for the CD version or the section of your website in which they can download their machine manual.”
But the future isn’t here, at least not for widely adopted, almost de facto standard industry HMI technologies.
“The problem with video is that the display quality on my HMI isn’t good enough for video. I suppose it doesn’t need to be in the HMI,” PAC Machinery’s Berguig thinks aloud. “I could use an iPad, and have videos pre-loaded, with a machine, for an extra $400."
Berguig is a strong proponent of, and adherent to, Moore’s Law, and expects video to be another frontier of HMI evolution. As manufacturers of controls differentiate themselves, more features will be available with the end user need and experience being the ultimate impetus. Souser at JLS continues to tinker and search for his own solutions to this demand.
“Its just awkward, though not impossible, with [Allen Bradley’s] PanelView, but we have to remember a lot of customers want that format,” Souser says. “But we did a video recently as a marketing and sales tool that shows how quickly you can change out tools. We could see video becoming the platform we use for service, and operator training on-board the machine. That’s when we have to start looking at other HMIs because we can't show that video well in PanelView.” Souser has used Beckhoff, and has considered B&R, citing speed, power and cost as positives, but comparatively low acceptance as a barrier.
“We have looked at other platforms, for cost and other reasons, but mainly because we were looking for that enhanced level of functionality, video etc. It's something you can get on our true PC-based machines, and you can’t get on PanelView without it being cost-prohibitive,” he says.
That said, Souser is no fan of Microsoft running his machines, so he’s also looking for some sort of environment that the machine itself isn’t dependent on. Looking back at his control-based difficulties over the years, PCs and Windows have been the number one culprit by an order of magnitude, he says. And he then gets blamed for Windows being Windows, and crashing or locking up. Meanwhile, Keller has less hesitation – he sees industrial PCs and PC-based control as the future, Bill Gates and all, and the HMI’s capabilities will flow from that.
“We follow customer standards, but always try to enlighten people on the opportunities, video or otherwise, that a Beckhoff, Siemens, B&R, and other players in industrial automation could bring. They’re at the forefront of the horsepower needed for the next 15 years. Whether that’s the Industrial Internet of Things, or HMI monitoring and providing real time outputs to mobile and the ease behind it, it means PC-based controls are the way of the future. In my opinion, the future will lean toward a single controller doing many tasks – operating flexible work cells.” Keller says.
While Beguig and Keller are turning over rocks and stones looking for more “horsepower” from their HMIs, Williams urges working directly with controls, component and HMI suppliers, and to establish a relationship with that looks more like a partnership than a vendor/buyer situation.
“We go to our supplier, whether that’s Schneider or Rockwell, and we are able to tell them, ‘this is what we need, this is what we are looking for, and this is what our customers are telling us that they need. And in order to operate this machine as effectively as possible, here are the things we’re going to need on that HMI,’” Williams says. “So we are constantly challenging them to look at the problem as a problem, not as something that has an existing solution.”
Going mobile – within limits
Mobility brings the industrial HMI world full circle. It takes the output screen of an industrial flat panel and drops it squarely on the very iPhone, iPad, or Android screen that industrial HMI’s have been chastised for lagging behind. But while making machine status changes from a grocery checkout line is doable, it’s a long way off from practical.
“There will have to be a proof of concept that it is very substantial for the end users in order to feel comfortable with true remote capability from an IT perspective,” Keller says. “When we go into plants, we are not allowed on the guest WiFi most times. And what does this architecture look like from a maintenance perspective on some of these remote monitoring devices? Is the world ready to support that once it goes live? I don't know if it will be as plug and play as your Apple TV at home.”
Souser does Keller one better when it comes to managing IT, cyber security, and end-user concern. “We have a customer who does not allow us, or their employees, to bring in any device that has a camera. Well, iPhones and iPads have cameras, so they’re out,” he says. It’s very limiting, and of course we can’t access that company remotely.”
Keller sees the remote HMI as having immediate function for alerting an operator, who might have an iPad at their desk while they are watching six machines so it's more of an output function. But he believes resetting or changing a machine state from a remote location is not in the foreseeable future.
Within the four walls is a different story. Keller and Souser both echo Berguig’s thoughts on using a secondary HMI device, as long as it stays within the building, and either can’t transmit out, or somehow is rendered as a local-only device. The question then becomes the expense of buying, then extracting functionality from a high-end tablet to render it good for a single purpose or set of purposes. Keller, Beguig and Souser are each independently toying with similar ideas.
“It's the primary and the secondary HMI idea – one is a troubleshooting tool for maintenance and operations which is mobile (at least within the facility), and the other is a primary unit of operation,” Keller says. “If you have a large cartoning machine that is 10m long, and you can only see the primary HMI from one point of view, both mobility and video capability in a mobile or secondary device comes in very handy. It helps with changeovers, it helps with maintenance and troubleshooting. That would be great. I think that should be the next focus for these types of integrated technology solutions.”
