A massive gearbox on an overhead crane catastrophically fails, sending the team into a reactive state of emergency repair and root cause analysis while leaders question the value of vibration analysis and remote monitoring they had been investing in to protect this asset… This is not the first time catastrophe has hit this crane, and they thought they were doing the right thing to protect it. So what happened?
A reliability engineer reports early signs of bearing failure on a major production motor to his team who are all avid believers in PdM. They all agree that immediate repairs are necessary and jump into action to protect the asset from failure, disrupting production, frantically pulling parts and documents together to get it repaired successfully, and celebrate when it goes back into service… only to be notified by the maintenance team after pulling it apart for rebuild that there was nothing detectably wrong with the bearing that was removed… so why did we replace it?
A maintenance superintendent reviews the vibration report from his contracted route-based service, and asks a few questions about the warnings and alarms noted. He gives direction to keep an eye on it, then goes back to business as usual, only to experience another catastrophic breakdown on the same assets highlighted in the report a few days later.
All three of these stories are true. You may recognize yourself in one of them. And it begs the question,
| Is PdM failing us?
The problem is not the PdM technology, and it’s not the application of the technology. The problem is the reactive mindset many organizations have when deploying it and the resulting programs that manage it.
After observing these situations across industry for years, we have come to believe that the attitude with which PdM is adopted and deployed has an enormous impact not only on what problems can be eliminated, but on how high your PdM program can grow.
How can something proactive in nature be used reactively in practice?
- Deploying PdM in response to a failure.
- Jumping into action at the first PdM alarm (without planning & scheduling)
- Waiting to act until it’s too late, even with a fault condition.
Let’s discuss these in more detail.
1. Deploying PdM in response to a failure.
Many times, the justification for using PdM comes from feeling the pain of a failure. The downtime from that critical gearbox failure was so large, we have to do something to prevent that from happening again! PdM can be deployed for a fraction of the cost of that failure, so it is now a justified cost!
In this case, the pain of failure justifies the need for prevention. PdM is applied as a reaction to a failure, so you may already be able to see that this is a reactive strategy for building a program for proactive tools…
Other than this being a reactive way to implement a proactive program, other problems may also arise if a complete understanding of the problem, and the capabilities and limitations of the technology are not clearly understood up front.
- Gear or bearing faults will be easy to find with the “rolls Royce” remote monitors, or the route-based analyzer, but some of the lower end models will be completely blind to these failures. Put a cheap sensor on a gearbox and when it fails, many will start to lose trust in the technology.
- Faults that are only evident under certain conditions may not show up for a route-based analysis, meaning you want to use a remote sensor and increase the density of data collected. Multiple “good” monthly vibration reports followed by a catastrophic failure will leave everyone wondering why PdM doesn’t work.
And maybe the bigger question is, “Why would we wait for a catastrophic failure to “justify” spending on proactive tools?” More on this later.
2. Jumping into action at the first PdM alarm (without planning & scheduling)
| PdM tools are intended to “offer you the ability to make different decisions.” – Levi
But when the PdM alarm goes off in the absence of other burning catastrophes, and we all jump into action to “fix it now,” it is a “catastrophic reactive decision” made in a situation that could remain calm and proactive.
The actions that follow a proactive fault signal should be very different than the call to jump into action for an asset that has had a catastrophic failure, but many times the response is the same, unfortunately.
3. Waiting to act until it’s too late, even with a fault condition.
Maybe even worse than jumping into action is “we have bigger fires to fight right now,” turning that catastrophic reactive decision making energy to all of the other things that are on currently on fire, ignoring the opportunity to get ahead of the even bigger one that is coming over the horizon.
| “Setup is 90% of the success of a job.”
Why is it so easy to get into this situation?
It is easy to say “hey, that’s the wrong way to do it,” but the reality is these types of decisions to deploy and use PdM in a reactive manner are so widespread because there may be some underlying issues driving these behaviors. In many cases, the reactive behaviors are inadvertently incentivized. Here are two examples.
How many times have you called out one of your technicians for being proactive?
| “Everyone congratulate Bob today! Bob saved the day by executing that PM inspection perfectly and identifying a condition that would have caused us downtime three months from now! Now we get to plan and schedule that repair so we will never miss production! Thanks Bob!”
More than likely, your praise is more often directed to heroes.
| “Bob came in at 2am and got us back up and running in less than 6 hours! Great repair Bob! You saved production this month!”
Most of us are conditioned to value the hero and “great feats of reactive repair” more than we value the quiet and calculated protection of function. It translates into our praise, and many times translates into other decisions we make.
Unfortunately, many times this mentality trickles from the top of the organization. Leadership, under pressure in increasingly competitive markets, looks for ways to reduce spending year after year. This has led many of us into only allowing spending when it is “necessary,” and since that failure didn’t happen yet, it doesn’t seem necessary. Small buckets of funding that could be used to reduce risk to production are allocated elsewhere, or cut altogether to help get under ever-decreasing spending ceilings. It is only when the catastrophic failure occurs that we may feel safer covering the enormous repair costs and deploying PdM as a prevention because it is easier to justify the spend once everyone sees that there is a greater spend if we do not.
This is not to criticize how budgeting decisions are made or how businesses are run, but just to highlight for those of us who are responsible for hitting those targets that we could be stuck in a reactive rut until we recognize it and make a conscious decision to “make a different decision.”
How do we get out?
Recommendations for the front line.
While it may seem very easy to build a reactive PdM program, it is not a trap that is impossible to escape.
- Identify your level of PdM program maturity.
- Consciously “make a different decision.”
- Build a proactive foundation.
1. Identify your level of PdM program maturity
The first step to getting out of the reactive PdM hole is to identify your current level of maturity. Here are 3 definitions of PdM program maturity for comparison to start.
WHACK-A-MOLE
Reactive PdM deployment in response to failure. Sure this may prevent another failure on that asset, but will it extend the life of the asset, and will it prevent failures on other assets that may impact the business even more than this one? Whack-a-mole PdM deployment is the most reactive maturity level of PdM programs.
SPRAY AND PRAY
Deploying PdM to anything and everything, casting a broad net, without strategic analysis of what and why. There may be a lot of faults identified and subsequent prevented failures, but there may be a lot of overspending of money and resources to get this accomplished, and there still may be some glaring gaps in preventing impactful failures or extending the life of assets. And your team may be exhausted from all the extra work running routes while still struggling to keep up with the corrective work! This strategy for PdM program deployment has slightly better performance results than Whack-a-mole, but is very expensive.
SURGICAL PRECISION
PdM is deployed carefully and in context of how it supports the business by reducing the highest risks for a palatable cost. PdM programs deployed in this context look at the business as a whole and use many of the reliability engineering tools to carefully analyze assets for the right opportunities for PdM balanced with the time, resources, and technologies available to the team.
2. Consciously ”make a different decision.”
Once you know at which level of PdM program maturity your team resides, step 2 involves making a conscious decision to “make a different decision” with the information you have available.
WHACK-A-MOLE
Take the faults you are receiving on the assets that are currently in your PdM program and make a conscious effort to plan and kit corrective work when it is identified. Before deploying more PdM to other assets, do a little work to ask not “is the failure painful and worth preventing,” but “is the asset and this specific failure mode important enough to us that it is worth preventing.”
SPRAY AND PRAY
Stop expanding your program and do a rudimentary analysis of the assets covered by PdM to see if paring back your program is feasible.
- Do some assets have multiple technologies covering the same failure modes? PMO is a great way to quickly identify assets that are “over monitored” and reduce spending quickly.
- Are some of your assets being monitored of so little importance that the business does not suffer at all if they fail? Are you continuing to do PM inspections for the same problems that PdM is supposed to inform? Discontinuing activities on these assets will help recover some wasted resources as you get ready for step 3.
Note: Planning & Scheduling are critical pre-requisites for effectively maturing a PdM program from reactive to proactive!
3. Build a proactive foundation.
Proactivity is built on the back of planning and analysis. Living in the future, not in the now. To mature to Surgical Precision in your PdM program, build a good asset hierarchy, apply asset criticality, then use FMECA, PPC, or PMO to determine how much time and resources should be allocated to reduce targeted areas of the inherent risk in your assets.
WHACK-A-MOLErs should look at expanding application of PdM to assets with the highest criticality as a next step, and using PdM to:
Find faults in the Predictive Domain of the P-F curve, and Move towards the Precision Domain of the P-F curve
After identifying the assets which have the highest inherent risk to the business, applying FMECA to those assets will further highlight which problems (or failure modes) have the greatest impact and what causes (or conditions) can be monitored through PdM to ensure function is adequately protected.
SPRAY AND PRAYers should look at paring down and focusing on the right problems. After the initial “paring down” in step two, using PMO (PM Optimization) is a tool that will quickly produce results from the efforts invested, and result in PdM being applied in more impactful and targeted areas while removing a lot of waste in general from the PM program.
Organizations that have applied PdM using SURGICAL PRECISION should continue to apply the good practices of continuous improvement through the work execution feedback loop, FRACAS, LEAN, regularly revisiting criticality and optimizing the maintenance strategy, or any other activity that reduces waste and improves results.
Recommendations for executives and senior leaders.
If you are in a senior leadership position in your organization, there are a few things you can do to help get your teams moving towards extreme cost reduction without having to ask for budget reductions. Knowing that spending is a result of the behaviors in practice at the front line, it is more important than ever to incentivize the behaviors that will lead to lower spending through process efficiency and improvement.
Tip: Spending is a result, not a behavior, so incentivizing the right behaviors is the only way to positively affect bottom-line budgetary impacts in the long-term without experiencing those “oh-no” moments of budget-busting catastrophes.
- Incentivize proactive behaviors. Taking a good look at the annual goal-setting to ensure your management team is incentivized to perform the behaviors that deliver results, rather than incentivized to deliver the results themselves. Look up Goodhart’s Law to learn more about this concept.
- Don’t wait for hard “losses” to approve proactive spending… Let your teams identify areas of elevated risk and approve or incentivize proactive programs like PdM which go after the top risks in your business.
Levi Heilman
Levi Heilman is Head of Solution Strategy at KCF Technologies, where he helps industrial manufacturers build and mature predictive maintenance programs using sensor technology and AI powered SMARTdiagnostics®. He brings years of hands-on experience as a vibration analyst and PdM service provider, having worked with thousands of customers across the United States. Levi specializes in bridging the gap between technology capability and real-world program execution.
Brian Hronchek
Brian Hronchek is Principal Trainer and Consultant at Eruditio, with 12 years of industrial operations and maintenance experience on the front lines of manufacturing and 17 years of leading teams through change in military environments. He works with organizations to build reliability-centered maintenance cultures, applying tools like FMECA, PMO, and RCM to reduce risk and drive sustainable operational performance.
