With dazzling wireless technology now available, it’s tempting for manufacturers to snatch up any wireless sensor that comes along as a means of optimizing processes and plant performance. This is especially true within the pharmaceutical industry, where vendors are plying industrial-strength wireless sensors for temperature, humidity and pressure, as well as sensitive process-monitoring wireless devices to support PAT applications.
But drug and other manufacturers may do well to exercise more caution when it comes to wireless. At least that’s the resounding message sent by a panel of experts who spoke on the current state of wireless standards at October’s annual ISA (International Society of Automation) Expo meeting in Houston.
The discussion drew an overflow crowd, and it was clear why the audience members, mostly from the end user community, had come: “Help!” If anything, wireless standards for industrial purposes have been getting more complex and confusing of late (see Box), and the audience was looking for a little comforting advice.
It did not come right away. “The wireless landscape is pretty confused, and will probably stay that way for a long, long time,” said wireless guru Hesh Kagan, Director of New Business Opportunities for Invensys Process Systems, near the start of the session. “Interoperability and coexistence are the issues we’re really grappling with, and I don’t have a prediction where standards will go, other than clearly they want to follow mass-market technology, which is driven by the communications and cell phone world.”
Jose Gutierrez, Corporate Director of Technology for Emerson Electric, agreed that telecommuncations companies are leading the advance of wireless technology, and cited what is being called Fourth Generation Broadband as a mobile, wireless technology that will have a transformative impact upon the wireless world in time. The problem is, as the panel pointed out, what’s good for commercial applications is rarely right for industrial ones. What’s a plant engineer with responsibility for wireless applications and planning to do?
Fortunately, the panel did get around to offering guidance and constructive advice, summarized here:
- Have a corporate wireless strategy. This strategy should be top-down and consider how wireless can work within the wired facility and within the company’s business model.
- Have a committee. Wireless decisions in one part of the plant can have far-reaching implications, and may even be antithetical to efforts elsewhere or to the business mission.
- Collaborate. In particular, get control and IT working together in decision-making.
- Start small. Try some wireless alerting and monitoring on a part of your plant that is not critical.
- Proceed cautiously.
Hail to the Incumbent
One of the problems with the wireless world of today is that, as in politics, the incumbent has the edge. Harry Forbes, an analyst with ARC advisory group, noted how the most accepted wireless (and wired, for that matter) protocols were also the oldest. People will stick to the same technology for a given application unless there’s a good reason for them to shift to a new one, Forbes noted. Thus for wireless, a Babel of standards has evolved.
Incumbency isn’t necessarily a bad thing, noted Ed Ladd, technical director for the HART Communication Foundation. “The old standards have migrated forward as the technology has improved,” he says. “These are evolutionary standards.”
There has been sufficient evolution to give most plant engineers the tools they need to build an effective wireless system, noted session moderator and Control magazine editor in chief Walt Boyes—this includes both industrial- and process-monitoring applications. “The landscape today is completely confused, as Hesh said,” said Boyes. “There’s all of these technologies, many of which are academic and not intended for industrial applications.”
However, Boyes noted, “there are three sensor network standards that you’re going to be seeing products from soon, in the next six months.” These include ZigBee, Wireless HART, and ISA100—though the ISA standards are not yet ratified, vendors are already developing prototypes for market based on ISA100.
“For purposes of what you’ve got to do as a decision maker, the standards are here,” Boyes said. “What isn’t here is the engineering tools to help you make those decisions. Companies need to develop those tools, and those are unfortunately behind the development of the standards.”
As others in the panel would do, Boyes cautioned engineers to develop their wireless strategies carefully and slowly. “Pay very careful attention and do it properly, or you will do rip and replace,” he said.
“Think about what you can deploy, how you can deploy, and how you can sustain it in the long run,” said Pat Schweitzer, Instrument Team Lead at ExxonMobil. “Because you basically can’t take it out of service.”
As Kagan put it, “The standard is secondary to the solution,” said Kagan.
What’s Your Strategy?
Honeywell’s David Kaufman, Business Development Director for Wireless Technologies, agreed that companies must be very shrewd in planning for wireless.“There are limitations in using wireless,” he said. “It’s not going to do everything you want to do, and you only have so much bandwidth. It time you will use all that up unless new bands open up.”
“I have talked to customers that have actually used up their bandwidth,” Kaufman said. These customers began to install wireless implementations and incrementally used up their entire band. “Were those the most important things that you wanted to use wireless for in your plant?” Kaufman asked the customer. “No,” the reply came. “They just kind of snuck in.”
Every company needs a wireless strategy, said Kaufman. “What’s most important for your company? What are the applications that you want to do? Wireless sensors? Fine. But do you want wireless handhelds, too? Fine. But in time are you going to have the bandwidth?”
“It’s a strategic decision, not a tactical decision,” said Kaufman. Forward-thinking companies are creating wireless governance teams, whose missions are to assess all of the potential wireless applications within their plants and to select those that are going to be truly impactful for the business.
As much as engineers would like to control decisions over wireless, in the end each individual application will have a businesswide implication. “Don’t go technology up, go business down and then look at what technology works for you,” Kaufman said. “It’s about tradeoffs. You can’t do everything.”
A company’s wireless model is not “flat,” echoed Schweitzer of ExxonMobil. “It should really be a 3-D model. Anybody who doesn’t take a look at this as a top-down problem will very shortly run into a problem with their network.”
A 3-D model is fully conscious of the company’s IT systems and requirements, Schweitzer noted. “You have to try to see a wireless landscape within a wired world.”
Wireless and Wired
A wireless strategy must be executed with a plant’s wired networks in mind. Wireless systems will never replace wired, particularly for control applications, said Ed Ladd. Wireless technology is advancing to the point that it may be able to handle some “soft” control responsibilities, he says, but there will always be limitations (such as data security issues) to wireless control.
You need to continue to use wired and wireless and implement whatever has the lowest business risk, Ladd added. “You will get immediate benefit from deployment of wireless technology,” he said, “but make sure you’re using it in the right way and you’re choosing what’s best for you.”
Current Wireless Standards
IEEE 802 (802.11, 802.15)—the most developed and popular (and oldest) of the standards organizations; “Everybody in the industrial supplier community is using 802.11 in some way,” says Harry Forbes of the ARC Advisory Group. 802.15 is the basic international standard for sensor networking.
ZigBee Alliance—a commercial consortium of vendors looking to capitalize and commercialize the IEEE 802.15.4 technology. After 2 or 3 evolutions, ZigBee has now gravitated towards the application of advanced metering. “All electric and gas meters in North America are going to be updated in next five to 10 years and they want to be part of that,” Forbes says.
HART Communication Foundation—a 15-year-old organization that has commercialized the incumbent standard of smart field devices.
IETF/IPSO Alliance—both promote the use of Internet protocol for sensor networks.
ISA—working on a number of different standards for wireless and seeking collaboration with other standards organizations.
TinyOS Alliance—standards used by academics who prefer an open source model that allows them to collaborate with colleagues around the globe
Bluetooth SIG—new implementation of Bluetooth (low-power Bluetooth) that will have traditional Bluetooth plus low-power application which may have some implementations in the sensor area.
Proprietary technologies (ANT, Z-Wave, others)