Get it right on paper before attempting to computerize and automate.
The Work Management Process. One of the biggest uses of IIoT, ML, and AI in the industry (so far) is in the field of condition monitoring and forecasting times to failure. IIoT devices deployed on your equipment and systems, send exception signals (they use edge computing to filter out the vast majority of the data that merely tells you “all is well”) over some sort of network (usually wireless), then some sort of software interface produces a message for you to read (maybe triggers and work request, or sounds an alarm) and then act on.
That last step – act on, is where many condition monitoring programs fall apart today even with lower-tech. If your work management process cannot reliably get someone out to investigate and resolve problems that are discovered, then the new tech won’t help. It will only add cost.
Your work management process (including field execution) needs to be working well. But it won’t do it without giving thought to how it should work. Get it right on paper before attempting to computerize and automate.
Your reliability program. What’s that you ask? You probably call it “PM”. It all those proactive maintenance activities aimed at ensuring reliable performance of your assets. Is that working well today? Can you get those PMs done on time every time? If not, then go back to “work management” and sort out some problems there. If you can, but you still don’t achieve reliable operations, then maybe the PMs are the problem. You need to look at those. Reliability Centered Maintenance (RCM) and related methods are probably needed for that. The tech won’t help with those either.
With the IIoT you can literally put devices just about anywhere. But you don’t need them just anywhere. In fact, you only need them in a few key locations and only to look at a few specific “conditions”. How do you know where and what to watch for? Again, RCM.
Look before you leap. With the CMMS and EAM systems of 25 years ago, many companies leapt before they looked. They learned a bit, refined their thinking, leapt again – perhaps a bit further this time. Most are still struggling with CMMS and EAM performance issues. And like I said, the problem is rarely the software.
With IIoT, AI, ML and digital transformation you can leap before you look. Some are doing that already. Some may even be getting encouraging results. Be more strategic and investigate first. That investigation isn’t just into the tech, as many are doing today. Look at your processes, problems, people, what can they do, not do, take on, etc. Get it right on paper before attempting to computerize and automate.
A recent case. We were working with a large and well-known company here in Canada. We did a quick assessment of maintenance performance in response to concerns that reliability was suffering. We found that it wasn’t suffering, but the way they measured it was. Related to that problem was a badly implemented maintenance management (software) system that just wasn’t performing. We dug a little bit and found that training was lacking (almost non-existent) and initial implementation was done on the cheap so the system was never very well configured to their needs. In fact, no one really knew the existing system very well (and it wasn’t all that old). They were considering its replacement. We recommended they put that on hold, engage the system supplier, investigate their configuration settings, training, and reporting needs. As it happens, the old system can do all they want, and more. With an investment of far less than alternative new systems would require, they can continue to use the system they already have and get the process performance and reporting that they need.
That same company is investing in a new plant that will be highly automated. IIoT devices are part of the design. They are trusting the engineering company that is designing and building it (a very reputable international firm) to get it all right for their needs. That is a lot of trust! We are now using RCM to define just what should be monitored so that the monitoring technology investment can be deployed so that it delivers value, not just a lot of signals. Along with that, we are redesigning how they do work management as we fix the configuration of the CMMS (above).
That integrated effort is being done to deliver what they need – reliable, high speed, and accurate operations in a specific window of time each and every day.
Conclusion: The tools are there, they are highly capable, and they must be deployed and used properly or you won’t get the results you think you’ll get. If you limit your focus to the tech alone, you’ll invest and get little in return. If you widen your focus to include the process and people, then you stand a much greater chance of getting those results and a high return on your investment in your assets.