3 reasons why PdM ROI often goes unmeasured

Predictive maintenance is only as valuable as its ROI.

By Sam McNair, P.E., CMRP, Life Cycle Engineering

In brief:

  • A PdM program should make solid returns on conservatively calculated and auditable repair costs that are avoided.
  • Plant managers often don’t know what the ROI is on their PdM programs because they don’t measure it.
  • Tracking PdM results and using those metrics to demonstrate effectiveness can help to cost-justify the program over the long haul.

Any good, well-managed PdM program should be showing a solid ROI. Running a program to make a profit for your company should be the rule, not the exception. One of the fundamentals of PdM program management I teach is: Show me the money. People too often get into PdM technology implementation because it’s a neat new thing or it’s a best practice or they have a talented guy on the staff or the place down the road does it, but without concrete plans to make money for the company out of it.

You might be able to get the seed money to start a program that way, but it’s no way to keep one. You can always show big savings if you can keep a gazillion-dollar-per-hour-profit production line from coming down due to a loose screw. But a well-managed PdM program can make solid returns just on conservatively calculated and auditable repair costs that are avoided. The downtime savings are just gravy.

Back in about 1997, I started putting all of my PdM and PM programs into a process where they kept key business metrics for their programs and published a quarterly P&L statement. It was a good way to run the programs and an excellent way to sustain the programs. You’d be surprised how often people have mature programs with well-trained people and good technology, but have a very poor or unknown return on investment.

Generally, there are three predominant causes. First it’s the result of inappropriate scope of application: I have a hammer, so everything is a loose nail.

Second and by far the biggest reason is that the findings of their programs go largely unused; it’s analogous to going to your doctor for a checkup, finding out your cholesterol and blood pressure are high, and then going home to eat potato chips and lie on the couch. The same goes for PdM program findings. I’ve surveyed a lot of sites and programs, and I’ve found that if as much as 30% of the things that are found are acted on within the recommended timeframe, you are considered average to good — only 30% to be at average or better. Now that’s a terrible use of resources. Have you ever wondered why PdM service providers don't publish statistics on how much day-in and day-out money they save clients? The big jackpot events get written up, but never any solid ROI or cash flow is shown. They tout percentages of assets in the green or percent of PdM coverage (a form of industrial keeping-up-with-the-Joneses advertising), but never the money. The reason is they are like that doctor. They can only advise, and they know that their clients in general don’t follow the advice. As a result, if anyone was watching ROI, which most companies don’t, they wouldn’t care to publish the results to the public because it would look so bad. There’s a lot of psychology of change, consumer education, good business process and good business management involved in making PdM a real profit maker for a company. An outside provider just can't do that. And although you can outsource the data capture and analysis, you can’t outsource managing your business processes for corrective action. You must own them.

Third and finally, most of the people on the inside who run the programs are technologists first, machinery people second and businesspeople not at all. I did an interesting case study on the utilization problem, first presented at the SMRP national conference back in 2000 and then presented publicly a few places after that. Here is an old sample of a real PdM program management mid-year report that shows how it’s done right. The money was based strictly on documented repair costs for the most conservative scenarios, audited by the accounting department, and blessed by the controller and the asset owners. The ROI is indisputably there. When the slash-and-burn people came by a few years after I had set this up, they left these programs alone and went on to cut programs that did not have their bottom-line contributions well-documented. Numbers like these make effective programs untouchable.

Sam McNair, P.E., CMRP is senior consultant at Life Cycle Engineering, headquartered in Charleston, South Carolina. Contact him at (843) 670-3784 or smcnair@lce.com.

Vibration Analysis Program
Management Statement 2003 mid year
  
Program Costs  
Software Maintenance Fee       $15,000
Direct Staff - 1/2 ME (loaded)                     $50,000
Vibration Techs - 2FTE (loaded)                     $120,000
Instruments Cailbration Fees/Repairs        $5,000
Training                      $4,000
Consumeable supplies                    $2,000
Allocated Costs $20,000
      Sub Total Costs $216,000
  
Value of Saves                     
Centrifuges                     $172,000
Pumps                     $186,000
Fans                     $80,000
Compressors                     $98,000
Mixers                    $62,000 
Sundynes                    $56,000 
Other                      $40,000 
      Sub Total Saves                $694,000
  
Financial Performance                     
Net Cash Flow/Day                     $1,330.63
Simple ROI                    62.40%
  
Performance Metrics/Data of Interest                     
% Discrepancy Compliance                    97.90%
% Route Compiance                    98.20%
Total Points Monitored                       40594
Points per man-hour                    53.4
Equipment per man-hour                    4.1
% Missed Points                    14.80%
Annualized Cost per unit Equipment                      $111.49
Average Cost per "Find"                    $749.67

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments