By Walt Boyes, Editor in ChiefIT'S VERY CLEAR
that if we want to have high performance manufacturing plants, we’ll need more and better sensors, more and better advanced process control algorithms, and better online calibration and maintenance capabilities than we do now.
The heart of this endeavor, without which it will be impossible, is wireless data transmission, and inexpensive wireless at that.
Proactive preventive maintenance is based on the idea that you can fix a motor, pump, valve or instrument before it breaks, and reduce catastrophic, unplanned downtime. However, predictive maintenance algorithms that are based on the statistical likelihood of a breakdown are absolutely defeated by the reality of point-source failure. You can predict that the likelihood of any motor train suffering a failure in the next 48 hours is x%, but you can’t predict that it will be the one critical train that runs product input to the plant that will be the one that fails.
"The SP100 committee is coming to ISA 2005 in Chicago to show how it can be done."
The only way to predict point-source failures is to monitor each point for failure parameters. It is far too costly to wire sensors to every motor, pump and device in the plant. Wireless, low power sensors are the option that makes sense.
But so far, that’s the only thing that makes sense.
As I pointed out in my article in the September issue ("Users Want an Industrial Wireless Standard
,” p. 103) we are in the thick of competing standards and companies all trying to manipulate the standards process to give their own products an edge. This is going to be essential when we start moving from asset management and reliability control to actual process control, and once the sensors and routers and transmitters are industrial grade, that point won’t be far away. See Dan Hebert’s "Into the Mesh
" column in this issue for an update (p.87).
If we have the same kind of catfight with industrial wireless standards that we had with industrial fieldbus standards, we will be guilty of shooting ourselves in the foot as a profession for the second time in a decade. We set back the ability to create truly integrated plant floor networks and efficiently move data from the plant floor to the enterprise a generation by arguing over which giant vendor would gain the advantage from the fieldbus standard. The end result is that there are now more than 120 “standards.” We did it, and we have to live with the results. Let’s not do it again.
I’ve put myself where my opinion is. I have joined ISA’s SP100 Wireless Standard committee, and I’ve started participating. SP100 is taking the role of coordinator for as many of the disparate standards bodies, including other ISA standards committees, which either have developed or are developing wireless standards. This includes the Hart Communication Foundation, the Fieldbus Foundation, WINA (the trade association of wireless network vendors) and IEEE. It remains to be seen whether ISA will be able to craft a workable, interoperable and completely non-proprietary standard for industrial wireless networks, which will effortlessly interoperate with existing enterprise wireless networks like Wi-Fi and Wi-Max and Zigbee.