The future of maintenance
This change in future is outside of maintenance. It resides in the education systems that feed our industries with maintainers and business leaders.
January 6, 2023 | By James Reyes-Picknell
Simply, maintenance is an important facet of keeping operations running reliably. To achieve that, we need to be smarter ourselves and work leaner, making greater use of technology deployed more strategically than we are today.
No engineer or inventor has ever designed an unbreakable system or equipment. Those who design our systems don’t see it that way, but for those of us in maintenance who do, it means job security. What we maintain and how we do it will change, but since we will be needed to maintain, is immutable. We already have much of the knowledge we need to do better, so a big change in future will be to leverage that knowledge. Rather than focus on developing even more complex and sophisticated systems, we will focus on using what we already have and what is already being developed.
We know preventing failures are good, and we know all failures cannot be prevented. We also know many failures (including some preventable ones) can be predicted if we monitor the right conditions. We can monitor equipment conditions (vibration, temperature) or observe system performance (fuel consumption, flows, valve positions, product quality). As maintainers we tend to do more of the former, rely less on the latter, and distrust operators when they complain about performance issues. We are sub-consciously filtering out valuable information.
Some conditions can be monitored by our human senses (low tech, high touch). Those tend to be inconsistent from person to person, but they can work if we watch for them consistently. Other conditions are monitored by instruments and equipment. Both require that we act when we find a problem.
Failures of condition monitoring programs are not usually failures of a sensor (human or otherwise), rather they are failures to be consistent in monitoring, and failures to do the needed follow up a problem is found. We already know this is a problem, with companies putting more emphasis on leveraging what it learns from its own operational information to better manage failure consequences than we are doing today. That requires some enhancement in planning and supervisory skills so that action is taken in a timely manner.
Ever since we have had computers, we have been experiencing massive growth in technology that enables better monitoring of equipment conditions. Today, it is becoming more sensitive, wireless, and in turn, sensor prices are dropping. The IIoT is enabling us even more, so our ability to observe is getting better all the time. Some systems are programmed to tell us what to do about those observations. It’s no different from predictive maintenance, except that it removes the human from the decision process. As machine learning improves that will grow.
For those who are willing to invest in monitoring, the future will bring us growth in what can be monitored, how we can monitor, and expand on remote monitoring. The use of robotics will expand along with the need to maintain those robots. That will come in the form of drones both flying and ‘walking’ through our plants and along our linear assets. Those drones will be bristling with sensors and readers to record what sensors installed on the assets are sending out.
Robots doing monitoring could potentially replace operators and maintainers who today are walking rounds. The robots will be more consistent, and likely more accurate. Readings will be recorded automatically, and more accurately than by humans. Warnings, and needed actions will be produced in some form of notification for action on our management systems. This exists today and there is a lot of growth potential here.
However, it will be incumbent on the maintainers to act on those notifications in a timely manner, or all that investment in technology will be a waste.
The knowledge needed to use condition monitoring technologies will become more embedded in the realm of AI and that specialized knowledge will reside only with the providers of those technologies. That will free maintenance workforce from the need to have that expertise. We will still need a maintenance workforce equipped and trained to replace components when they are failed and to perform troubleshooting that goes beyond what the sensors tell us, and to confirm what the sensors are telling us. Maintainers will become better trouble-shooters, or the parts inventories needed to supply guesswork parts replacements will become very expensive.
Taking full advantage of that requires the knowhow of what can go wrong and what to look for to monitor effectively. It requires a deeper knowledge of what is happening inside our assets – the sort of knowledge generated when we do reliability centered maintenance (RCM) analysis. The future will bring more of that. Since it is so sensitive to operating context, we will learn that one size doesn’t fit all and resistance to investing will fall.
All this means that we will find failures sooner, minimizing business consequences and likely doing less extensive repair work. That repair work will still require parts and materials, all available when and where needed. Historically we have done, and continue to do, a very bad job at forecasting those needs.
There is an emerging awareness that we can leverage RCM results that tells us what can happen, how often, and what we can do about it, to enable better material needs forecasting. Without that, we will continue to suffer materials shortages and deferred or delayed repair work. One solution is to build a more robustness and more redundancy into our assets, but that is very capital intensive. A better solution is to get smarter and leverage what we know works well.
There will be continual introduction of new technologies into our physical assets – plants, and mobile equipment. As we introduce those, we will need to learn how they work (so we can operate them) and how they fail (so we can maintain them). Some will dictate new maintenance methods that will also need to be taught.
Presently, many companies loathe to invest in anything but the minimal training to meet regulatory requirements. If we take advantage of these sophisticated new technologies – then training must include them. As essential as this will be, maintainers will likely continue to be poor communicators of needs and business benefits, and companies will continue to lag in their investments in the training needed to leverage technologies most effectively.
Technical people need to improve communication skills and business knowledge. Business leaders need a better understanding of the assets they depend on – they can meet in the middle and communication challenges will be less. Eventually business schools and technical education programs will dedicate some of their curriculum to these important subjects that today are largely ignored. This change in future is outside of maintenance. It resides in the education systems that feed our industries with maintainers and business leaders.
James Reyes-Picknell is President of Conscious Asset and the Author of Uptime – Strategies for Excellence in Maintenance Management (Productivity Press, 2015). Reach him by phone at 705-719-4945, e-mail him at email@example.com or visit www.consciousasset.com.