All Member Open Forum

We have three significant dates for our Preventive Maintenance - work order creation date, scheduled date and due date. Probably the same as about everyone. Debate in my company is that unexpected events happen which should allow moving the due date out. I have guidance on the date range of completion for each PM interval, such as 90 days for an annual which should be more than adequate.

Besides making your PM compliance KPI better, struggling with the purpose of moving the due date out. Just document and get done when able.

Does anyone move out PM due dates?

------------------------------
Lee McClish
Manager, Maintenance and Reliability
NTT Global Data Centers Americas
Ashburn VA
------------------------------

I have only moved out metered PMs.  We try to estimate when production with hit their targets and convert meter to an estimated date. But some lines do better and some worse.  You have a +/- 10%. So you have to work with operations on scheduling.

------------------------------
David Skarupa
Reliability Manager
Domtar Personal Care
Columbus OH
------------------------------

The schedule doesn't always work as planned which is why you have a compliance window for PMs. This ideally is 10% of the frequency. This would be 3 days 30x10%) before or after on a monthly (6 day window- 3 before and 3 after) . That said PM completion is usually higher than Compliance because some miss the established window but are still done. That's ok but you should try to optimize your frequency. Is there an impact for late or early PM and are those PMs finding defects or future failures. Pending that answer, increase or reduce frequency duration and or modify the inspection task. The key point of a PM program is "preventing" unplanned down time. Optimize you PMs to achieve that goal. Your PM KPI should identify opportunities in this area. Hope that helps.

------------------------------
Jody Witham CMRP, PMP, STS
Dallas GA
------------------------------

We follow the 10% rule as explained in other replies but the biggest window I use is 30 days, which should be adequate even in our 24/7 facility.  If PMs are missed we count them against our PM completion KPI and get them out of the system.  The only ones we keep active, outside the window, are regulatory type or PMs that we know will definitely cause issues if they aren't done.  Those stay open until we can get at them done but are counted against our "on-time" completion KPI.  My view, if you are not completing your PMs in the specified time you have defined, you don't have a good PM program.  While all of the PMs may not get completed within the window, you need to have the data to improve.

Also, if you miss some PMs just move on and get them done the next time they are called.  If you keep them all open you are causing a snowball effect that you may never get caught up while leaving your CMMS in a mess.  You might end up realizing you are overdoing PMs if they are getting missed and not causing issues.

A good thing to have is a PM calendar that lists what machines you have scheduled and the frequency of the PM scheduled for the year.  Then everyone can see what is due and communicate ahead of time.  We still make adjustments but try to keep them in the window.  If we cant keep the timeframe and still do the PMs they are all counted against our KPIs but now we know the reasons and can improve in the future.

------------------------------
Brad Markert
Amcor Flexibles North America
Winneconne WI
------------------------------

I've moved PM due dates (some forward and some back) to avoid a resource crunch during scheduled outages - months ahead of time. If the idea is to push the due date back by a week or two because of unplanned events, then I'd agree that it feels like a a way to play the KPIs. Wouldn't that make the KPI invalid? A dishonest KPI can't help the organization improve - which is the point of tracking KPIs, right?

Looking forward to other's point of view on the topic. Great question, Lee.


------------------------------
Daniel Ferrell CMRP
Electrical Reliability Specialist
Flint Hills Resources, LLC
Corpus Christi TX
------------------------------

Totally agree!  If your CMMS is already in a mess then it's best to reset things and start fresh.  AFTER that one reset, you need to hold yourself accountable via your KPI's or otherwise your metric is useless.

------------------------------
Angela Wall
Senior Manager Engineering & Maintenance Operations
Butterball LLC
Mount Olive NC
------------------------------

Original Message:
Sent: 10-29-2021 09:06 AM
From: Daniel Ferrell
Subject: PM work order due dates

I've moved PM due dates (some forward and some back) to avoid a resource crunch during scheduled outages - months ahead of time. If the idea is to push the due date back by a week or two because of unplanned events, then I'd agree that it feels like a a way to play the KPIs. Wouldn't that make the KPI invalid? A dishonest KPI can't help the organization improve - which is the point of tracking KPIs, right?

Looking forward to other's point of view on the topic. Great question, Lee.


------------------------------
Daniel Ferrell CMRP
Electrical Reliability Specialist
Flint Hills Resources, LLC
Corpus Christi TX
------------------------------

We all know that PM addresses known failures modes. Which should have been identified when an FMEA was done before install and commission. There are many times when we do have to push out PM due to unforeseen circumstances. That is why we are pushing to get our PM done, not on time based but meter based frequencies. The question is, if you do not do the PM to address known failure modes, the failures will happen. That is why we also have two other KPI's and they are PM completed per plan and PM effectiveness. The first ensures that we are doing the PM consistently and with little to no variation, or does the plan and total cost match the actual after completion. This is measured by a target of +-30% of plan VS actual. On the KPI PM effectiveness we measure how many follow up jobs are generated from a PM. Are we actually finding items that are not functioning as intended. To get the best useful life out of assets and components we want to find that issue right before it fails far enough to cause a functional failure. We also use this KPI to make sure we have the right duration between PM inspections. Unless it is a compliance issue, why do we check a bearing every month if we never find an issue with that bearing. The PM effectiveness KPI has a target of 2%. Which means that 2% of all PM work that is done, we find an item is not functioning as intended. I know that 2% is low, but we are trying to get all our maintenance managers to understand the concept before we raise target %.

------------------------------
Phillip Swanton RE, CRL

Honda Development and Manufacturing of America, LLC

Production Engineering Business Unit/Manufacturing Strategies & Standards Department

Technical Development Unit/Asset Management

------------------------------

Due dates, by definition, are not flexible. They are the dates which the organization has set to comply with the designed planned maintenance program. If someone wants to push out completion past a due date, then a documented risk analysis should be completed (complexity varying on the importance of the asset and failure mode being addressed) and approved prior to the push. The ability to continuously miss due dates with no negative results is an indication that the PM program may not be fully justified. When PMs are missed and nothing bad occurs (or nothing bad happening in the future can be traced back to the missed PM), we reinforce the idea that planned maintenance completion is optional both with Operations and with our Maintenance department. Only by understanding failure modes and tying those failure modes to the maintenance routines can we truly justify why we do what we do, and when we do it. I know I got a bit off the topic of moving due dates, but at the end of the day, we aren't managing due dates, we are managing failure modes to maximize availability with minimum cost.

------------------------------
Mark West CMRP, CRL
Director, Reliability Services
ABS Group
McKinney TX
------------------------------

Hello Lee,

When scheduling planned work, do you consider Work Center capacity? What we have found is that no matter how good your maintenance team is, the chances of required unscheduled, corrective maintained are always present and more often than not, do occur. If you do not make provisions for accommodating this probability, your chances of having a schedule breaking event are high. The approach we take is to schedule Work Center capacity to something less than 100%. Maybe 70% to 80%. If unscheduled, corrective maintenance does occur AND if it is of an urgent or even emergent nature, you will have capacity built into your near term look-ahead schedule to accomplish this work without the need to push out your locked-down Work Order schedule past the time your originally intended. In other words, you may not have a schedule breaking event. On the flip side of this scenario, if you find that after many months, your Work Centers have continuous lag in their schedule, you can increase capacity incrementally by a small amount Or possibly pull in work from your unscheduled backlog forward mid-week to fill the Work Center capacity gaps. The goal is to sustain your schedule going into T-1 and definitely T-0 weeks and keep your Work Centers wrench time at expected capacity.

------------------------------
Ed Espinosa, CMRP, CRL, PMP
Sr. Performance Analyst
Puget Sound Energy
Bellingham WA
------------------------------

Lee,
I completely agree with Mark West's response. Discipline to the definition of metrics is critical. Otherwise the message being sent is everything is flexible so we can make things look better than they really are.

Additionally, the root of the problem presented is variation in operating systems. You don't improve PM performance by adjusting due dates. You improve the performance by improving operating system stability. A great opportunity to engage with operations on defect elimination to remove waste and inefficiency.

------------------------------
Tom Moriarty
President, Alidade MER, Inc.
Satellite Beach FL
http://www.alidade-mer.com
+1 (321) 773-3356
tjmpe@alidade-mer.com
------------------------------

Your scheduled PM dates should not be moved for simple convenience (Or for any reason, for that matter). If you move the dates, as others have said, you send the message that the schedule is not important, and this undermines the whole PM program.

There is a reason for regular, systematic inspections, and changing or moving the due dates makes a hash of the program.

In a previous life, if the PM was not performed in a timely fashion (established and communicated to all) it was canceled with a justification noted.

Everything we do should be used to drive improvement and reliability, even our failures! When we do not accomplish our goals, this needs to be documented, and used as an opportunity to improve (Not just point fingers). If the machine that did not get inspected, experienced a failure that should have been caught on the routine inspection, this needs to be communicated, and used to further support the program.

------------------------------
David Rempel
RELIABILITY PROFESSIONAL
Allied Reliability Group
Houston TX
------------------------------

Lots of good answers already @Lee McClish

I agree that in principle due dates should not be moved, otherwise you slowly create a culture where people focus more on managing the metrics than doing the work. 

That said if you have a major outage and that has been shifted by a month and now the due dates of a lot of your PMs fall before that outage it does make sense to update your due dates rather than have all of those tasks go as late completions. But if these are critical PMs (e.g. legislative / safety-critical) you want to risk assess that and make sure you have some kind of approval process when changing due dates. 
​

------------------------------
Erik Hupje
http://www.roadtoreliability.com
https://www.linkedin.com/in/erikhupje/
------------------------------

PM due date should be fixed and completed in the agreed compliance windows (10% of the frequency). Major turnarounds or others should not impact them as there is a specific reasons why a certain frequency (calendar or usage base) was set.  We only adjust the schedule dates base on existing constraints. When adjusting those dates, we should assess the risk of not completing the PM or PdM as per the due date. KPIs will track compliance base on the due dates and we need to be able to explain the lower value and potentially breakdown associated with the delay.

------------------------------
Charles Henault
ISPE Reliability Manager
Orica Ltd
Brownsburg-Chatham QC
------------------------------

Original Message:
Sent: 11-04-2021 01:40 AM
From: Erik Hupje
Subject: PM work order due dates

Lots of good answers already @Lee McClish

I agree that in principle due dates should not be moved, otherwise you slowly create a culture where people focus more on managing the metrics than doing the work.

That said if you have a major outage and that has been shifted by a month and now the due dates of a lot of your PMs fall before that outage it does make sense to update your due dates rather than have all of those tasks go as late completions. But if these are critical PMs (e.g. legislative / safety-critical) you want to risk assess that and make sure you have some kind of approval process when changing due dates.
​

------------------------------
Erik Hupje
http://www.roadtoreliability.com
https://www.linkedin.com/in/erikhupje/
------------------------------