The process-based wireless sensor networks (WSN) WirelessHART and ISA-100.11.a have been on the market for more than seven years, yet true to form, most facilities have not yet fully adopted them. I suspect many are still from Missouri, the “Show-Me” state, as in, “Show me in someone else’s facility the exact application I’m considering running.”
Being engineers, we’re averse to risk, and because we rely on sensors to keep our facilities within safe operating conditions, we need to know wireless works before installing it in more than monitoring applications. Even then, we need to be careful of which applications because if we start measuring, we’ll also have to report it if asked. However, it’s somewhat ironic that the majority of users are still in this mindset because ISA-dTR84.00.08, “Guidance for Application of Wireless Sensor Technology to Non-SIS Independent Protection Layers” is presently with ISA-84 for ballot until July 11.
The HART 7 specification defining WirelessHART was released 2007, with the first product from Emerson in September 2008, while Yokogawa released the first ISA100.11a products in September 2009. So we can safely say products have been available for more than five years. Looking back at Fieldbus days, WSNs are at about the same level of acceptance, and it took about 10 years for folks to get to the “this stuff works” mindset. I say this based to some extent on what I observe in the market. When I approached the consortia supporting these technologies, they were unable to provide information on installed base, though they did confirm that each technology is dominated by one supplier. During the fieldbus wars, they were trumpeting their market shares, so the good news is that the wireless wars, if they exist, are being fought differently this time.
One place where I was able to get some information on wireless adoption was from On World’s “Enterprise IoT Survey,” inaccurately named because it was all about wireless, though the majority of pundits agree wireless will be a critical part of IoT in whatever shape it evolves. Some of the interesting statistics from the survey were:
- WSN protocol usage had 802.11 at 38% (I don’t know of many process WSN devices using Wi-Fi), but WirelessHART was at 18% and ISA100.11a at 14%, so they were pretty close to each other. (This question asked if you used the technology, but not how many devices were installed);
- One in three use mostly mesh networks, but one in four have no mesh nodes;
- 30% of wireless sensors have some form of energy harvesting; and
- The most commonly named future applications for WSN technology were environmental monitoring (49%), asset monitoring (45%), process monitoring (41%) and then, much lower in the rankings, process control (19%).
Remember that process market is but one small fraction of the total wireless market. This appears to be reflected in the survey numbers.
The survey also asked what is the largest impediment to greater adoption (lowest satisfaction), and respondents indicated battery life, cost and integration caused the most grief, while the most highly ranked important features were reliability, security and cost. The next set of data showed that some folks are getting their sensor nodes for less than $50 (obviously not the process market), while approximately 20% of the respondents are paying more than $1,000 (probably the process market).
You could likely see a $50 node if you looked at the light poles on your block, and noticed that one of them has what appears to be a Wi-Fi access point for the local utility meter reading system (no power problems there).
All these statistics have me wondering just where we are today on the acceptance curve. I certainly know it’s not the plateau of productivity—unless, of course, you know otherwise and are willing to “take me to Missouri.”