Special Section: Wireless
Beauty and the beast
Users want happy union with wireless, but doubts linger
By Patrick Schweitzer
When it comes to wireless technology, the user community wants simplicity, robustness, education to understand this new world, and coexistence among all players to make this the wireless world of our dreams. Yet as wireless picks up momentum in the industrial marketplace, users are still confounded by the potentials and pitfalls a wireless world can bring. Ironically, they also make up the ultimate partner that will develop and deploy the visions suppliers bring to the story. To make this fairy tale a dream come true, a paradigm shift needs to occur in the user community so they can gain the same level of confidence already established within wired infrastructures.
What users want
Users want a level playing field where they can achieve interoperability and coexistence at various levels. They are also looking for guidance in managing the wireless space, and although they may need some specialty solutions, they want a common network solution that encompasses the overall landscape. Does the solution need to be in one box? Technology has the inertia to head in that direction, but present day users are expecting a cohesive and manageable network adaptable to the plant environment.
Users know they can only achieve the potential for large gains in productivity and efficiency with wireless as an option. The key word is simplicity. The networks initially need to mimic the wired environment. This will aid transition as we retrain our frontline professionals in the technology. But replacing copper only grabs the low hanging fruit. The real tipping point will happen as we reshape conventional applications.
Although the potential for numerous applications exist, it can be a daunting task to manage the environment. Take a look from the top down. It is never about the first application installed. It takes forethought that other installations will be built on the success of the first. Starting with a main backhaul network, high bandwidth users need to be managed alongside lower data rate sensors. Management and separation at network levels to insure availability and security becomes a dynamic task wirelessly. In the wired world, separation was a physical issue and easily understood. Users are looking for parallels in wireless networks that achieve the same goals.
Should we risk it?
The next issue is risk. ISA100 defined use cases encompassing the plant data environment. But in a system built on collected failure data, wireless does not have the pedigree wired systems have. The rest is a matter of managed risk in the process, and each user will struggle to decide the pace at which they will adopt each of the use cases.
The other risk is security. An IT professional views security of a network differently than a control system network professional. The combination of the two definitions can make a wireless signal non-deterministic in origin. Some groups view the wireless connection as external or, simply put, outside the complete control of the network. Users are looking to the supplier and technical community for help in definitions and testing protocols that are verifiable and repeatable.
Undefined security is another roadblock. Although the supplier community is touting security at a level greater than a wire infrastructure, users are still skeptical. Outside agencies and hackers are daily finding ways to infiltrate the seemingly safe system. This makes the issue dynamic, which in a production environment is sometimes unacceptable. Suppliers are turning to agencies for security testing, but there is no standard that dictates testing and is repeatable for users. Users therefore are requiring clear and concise documentation on security testing and protocols that clearly define the issue and can easily be proven at site.
Security needs to meet online requirements of the working plant. IT views security in terms of CIA, that is, confidentiality, data integrity, and availability. For IT, denial of service or a system blockade is an available weapon to thwart attack. This will not work in an operating facility. The network needs to be as close to 100% as possible in terms of sensor and network availability.
Are we compatible?
Compatibility with wired networks is also an issue. In a world where business and control networks are kept separate and data is carefully controlled as it is passed up from control to the business layer and never the other way around, wireless can shake up that paradigm; both networks are sharing the same airwaves. Even though some can prove otherwise, the IT and control factions are still viewing wireless as non-deterministic and therefore an external connection to the wired network. Users need education here to help bridge this gap before adoption will happen universally.
Other stigmas slowing adoption are numerous and conflicting standards, in which frequencies are hard to understand and manage for the new user. Growth seems to be haphazard, with one-off applications cropping up and no plans for a consolidated and integrated network. Cost is also an issue since only the capital avoidance part is readily quantified. New applications will take new studies to determine real savings, efficiencies, and improvements. Documented, new applications are slow in appearing in the literature as new users and suppliers alike are working out the bugs of the new technology.
Expertise in the user community is another issue. We have a high level of network knowledge, but in most cases, a low level of radio knowledge. Education will become a big factor in the success of wireless. Although millions of smart transmitters exist in plants today studies show diagnostic information still is relatively under-utilized. Although wireless can free that stranded data, is that enough to get users to realize the potential of the information? In most cases, it is lack of understanding of the information available and how we can use it.
Reaching a compromise
The user’s vision in reality is a coexistence strategy that encompasses wired and wireless infrastructures. It also spans the strategies that exist in both the business layers and control layers. It becomes not an issue of who owns the network, but how we can live in the same space. Wireless opened up a large, gray space that still has uncertainties for users. They are faced with proving the robustness of the new technology in terms only understood in a wired environment.
In some cases, there is no translation, and it is a matter of a paradigm shift in understanding control and IT organizations. The deployment and ultimate growth depends on a concise plan with buy-in and understanding from all parties. Those parties though are requiring proof they can reproduce in various environments. They also need to know what the changing technology will do to the operating system once installed. In essence, they are looking for future-proofing of the wireless network. That is the challenge. It is not without risk, but it has a far-reaching potential.
Wireless opens up opportunities for unique applications. Take data stranded in smart transmitters. What benefits have we gained from the present use of HART diagnostics? Would more extensive training on what is in the HART protocol extend the use of that information?
As far as monitoring equipment, now wireless sensors only scratch the surface of what users would like to do. With the present battery life and radio technology, EHM data is, for the most part, static. Technology is progressing to where dynamic data could be possible allowing measurement in real time.
Users also want low-cost sensors. How about a low-cost valve monitor that would allow a plant to know the position of all the valves in a unit? If you look back at the recent history of incidents in the process industry, in almost all of them the root cause leads in some respect to a manual valve issue. Wireless could bring in a new level of safety to our facilities.
We could start looking at self-organizing networks, designed to organize themselves in response to a request or a problem. From there, we could gather data and view it in such a way to make the optimal decision whether it is more production or less energy usage. We would view the data differently depending on the person screening it. An operator could make a decision to move a set point, or a venture manager could view possible increased output—all of this from coordinated networks that respond to the request.
ABOUT THE AUTHOR
Patrick Schweitzer is team leader in instrumentation measurement and automation projects at ExxonMobil Research & Engineering in Fairfax, Va. Contact him at firstname.lastname@example.org.