All Member Open Forum

 View Only
Expand all | Collapse all

PM Backlog

  • 1.  PM Backlog

    Posted 01-22-2020 06:10 PM

    Hello all,

    I was looking for some opinions on what your thoughts and experiences are with PM backlogs.
    The plant I work in is a single train petrochemical production line that has a historical issue of backlog PM routines. There is some redundancy built in to the plant but not a lot, and some of the main pumps we have cannot be swapped over on the run, or the operators are not willing to.
    I am new to the plant and when the planners asking what the issues are with backlogs, I was told it is mainly due to PM's cannot be completed as awaiting plant condition, I.e. a shutdown as there is no proper way to do the majority of these PM's on the run.

    The issue seems to be there is a lot of reactive maintenance which obviously does not help so emergent work during running periods then gets prioritised before PM routines - and the cycle continues.

    The management are now posing the question of if we could utilise contractors during shutdowns and outages to do the bulk of low skilled PM routines. Has anyone done this and what were your experiences?

    For me the current routines are poor, with many time based routines and not a lot of condition monitoring (plant is 20 year old) and we should be shedding low value PM routines and moving as much as we can to CBM for equipment.

    Appreciate your thoughts and experiences on PM backlogs and how you over came it.

    Thanks
    Derek Brown 



    ------------------------------
    Derek Brown
    Grangemouth AB
    ------------------------------


  • 2.  RE: PM Backlog

    Posted 01-23-2020 09:40 AM
    Derek, it sounds like you're caught in the old Fram oil filter cycle - pay me now or pay me (a lot more) later. If I were in your situation as the new guy, here's the approach I would take:
    • Examine the current PM content for validity, because the PMs that require downtime need to be as short of a duration as possible. Try to find the technical basis for each of the tasks and tie them back to the failure modes they are intended to prevent or detect. If a task can't be tied to a failure mode, it's probably not worth doing. 
    • Do an FMEA to ensure you are covering all the failure modes that you anticipate.
    • Address as many of the failure modes as you can with condition monitoring technologies. 
    • For those tasks that require downtime, try to find the right frequency by examining equipment history and/or talking to the experienced maintenance folks. 
    You will now have an "optimized" maintenance strategy that is absolutely necessary to be executed to protect the equipment. Failure to do so will most certainly result in equipment failure, and you have to get management support to get the downtime needed. Those PM tasks need to carry equal priority as the corrective tasks, so the resources have to be dedicated to their execution. Take advantage of any downtime windows from other emergencies to get them done. 

    Hope this helps!

    ------------------------------
    Bruce Hawkins
    Director, Technical Excellence
    Emerson Process Management
    Pendleton SC
    ------------------------------



  • 3.  RE: PM Backlog

    Posted 01-26-2020 07:55 AM
    I completely agree with Bruce's advice, but would add that if you determine that certain PM tasks are valid for mitigating a failure mode,  you also need to determine the proper interval for that task.  So, how were the original intervals selected?  Probably just guesses, past history or perhaps an OEM recommendation.  Your post doesn't mention if you are seeing equipment failures that might be traced back to NOT performing some of these PMs.  If there are not failures related to those tasks once you do the FMEAs, then certainly the interval between the tasks could be increased.  

    If there do appear to be failures related to a particular failure mode, you may be able to do some Weibull analysis and determine better intervals for these tasks.

    ------------------------------
    Joe Petersen
    Consultant Trainer
    GE Digital
    Knoxville TN
    ------------------------------



  • 4.  RE: PM Backlog

    Posted 01-26-2020 04:24 PM
    Bruce gives the best advice. You’re looking for the nearest alligator to your canoe. As Bruce and other have pointed out, if you’re missing PMs, it might not matter if they were useless to begin with. Or, doing them is actually an opportunity to insert a fault.

    ---------------------------------
    Karl Burnett

    Solvay
    Anderson SC
    ---------------------------------





  • 5.  RE: PM Backlog

    Posted 01-31-2020 10:00 AM

    Bruce,

    Thanks for the detailed reply and agree with everything you have mentioned.
    However, as afore mentioned with the plant being as old as it, is it worth the effort of going through the detailed analysis you mention above? I am playing devil's advocate here. Unfortunately we are 20 years too late. 

    A plant is only as old as management allow it to be, but typically once they reach 20+ years and start to become harder to maintain, this is usually the beginning of the end (in most cases).

    Is it the case of regardless of how old a plant is it shouldn't matter? 

    As an engineer is goes against everything I've been taught when it comes to maintaining plant to continue to run as is. But it really needs to be driven top down from management, and I am unsure what the appetite for this change would be.

    Thanks
    Derek 





    ------------------------------
    Derek Brown
    Grangemouth AB
    ------------------------------



  • 6.  RE: PM Backlog

    Posted 02-03-2020 08:39 AM
    Derek, I worked at a 30 year old plant that made a nearly 10% jump in OEE while reducing maintenance costs by over 50%, so it can be done. I suspect that the PMs were not getting done because the plant staff (both maintenance and operations) believed they add no value or were not important. That's why I suggested that you go through the engineering analysis first to take that argument off the table. 
    You are correct that it needs to be driven from the top; a reactive culture is really hard to change. I would suspect that you were hired for just this purpose, but as a new guy, folks may not readily listen to you. The best way to get the ball rolling is to demonstrate some success so you gain some credibility with management, even if it's on a single asset. In addition, it would be helpful to develop a business case to illustrate how much money is being left on the table in the current operating mode. I can help with that if you'd like. My email address is bruce.hawkins@emerson.com ; and good luck on your journey. It will be a difficult one, but well worth it.

    ------------------------------
    Bruce Hawkins, CMRP, CAMA
    Director of Technical Excellence
    Emerson Automation Solutions
    Bruce.Hawkins@Emerson.com
    ------------------------------



  • 7.  RE: PM Backlog

    Posted 01-31-2020 10:05 AM

    Bruce,

    Thanks for the detailed reply and agree with everything you have mentioned.
    However, as afore mentioned with the plant being as old as it, is it worth the effort of going through the detailed analysis you mention above? I am playing devil's advocate here. Unfortunately we are 20 years too late. 

    A plant is only as old as management allow it to be, but typically once they reach 20+ years and start to become harder to maintain, this is usually the beginning of the end (in most cases).

    Is it the case of regardless of how old a plant is it shouldn't matter? 

    As an engineer is goes against everything I've been taught when it comes to maintaining plant to continue to run as is. But it really needs to be driven top down from management, and I am unsure what the appetite for this change would be.

    Thanks
    Derek 



    ------------------------------
    Derek Brown
    Grangemouth AB
    ------------------------------



  • 8.  RE: PM Backlog

    Posted 01-23-2020 09:48 AM
    We currently use Contractors to perform the Vast Majority of our scheduled Shutdown PM Activities and, from my standpoint, it has proven beneficial.  It frees up our skilled maintenance personnel to work on specific issues.  We also rely heavily on our Reliability Engineers to try and determine root cause for any issues we have then, assign our personnel to correct them per the RE's recommendations.

    ------------------------------
    Jack Webb
    Maintenance Planner
    United States Sugar Corp
    Clewiston FL
    ------------------------------



  • 9.  RE: PM Backlog

    Posted 01-23-2020 09:59 AM
    Hi Derek,
    Accumulated backlogs are real concern and it has to be tackled by revising Maintenance strategy. Based on criticality, PM's can be fine tuned. Shutdown T/A is for major activities and it should not be wasted for PM. A proper condition monitoring and predictive maintenance can help until you revise your strategy.

    ------------------------------
    Mugundan Devanathan
    Mechanical Supervisor
    Abu Dhabi Gas Industries Ltd.(GASCO)
    AUH
    ------------------------------



  • 10.  RE: PM Backlog

    Posted 01-23-2020 10:01 AM

    Hi Derek, 
    You ask a question about a very common situation we see especially when the reactive (I.e. breakdowns) maintenance is high. 

    When we see high levels of reactive maintenance we also see a corresponding low compliance in executing scheduled PM work (I.e. PM backlog is high). So the question is why?

    This is where you find the root cause(s) ..... 
    1. Is there also a poor routine in the Planning and Scheduling of the maintenance labor? 
    2. Is the quality of the PM program poor? Sounds like you believe this is true.
    3. What are the beliefs within the organization (I.e. Leadership) about Pro-active maintenance? If #1 is yes, it is a clue on the answer to this question and where the current situation stems from.

    Steps towards a Solution - 
    1. Conduct a full-blown assessment on the overall situation at your facility. Only then will you be able to understand what to do to resolve the problems. The top 3 key areas of the assessment - Leadership, P&S and PM.
    2. From what is found in the assessment, build a plan with very specific tasks on what to do to close the gaps toward improvement. Part of this would include a deep dive in to the specifics of your current PM program. 
    3. Prioritize the tasks and make assignments on responsibilities.
    4. Execute and measure!

    This may all sound easy but it is very difficult because it involves people and changing the status-quo. 

    The  poor results of your PM program is a symptom of a more systemic problem. You must go after the right problems for a long term and sustainable solution. If you do this, your PM program gets much better along with many other components of a good work management program.



    ------------------------------
    Terry Taylor
    Senior Consultant
    IDCON INC
    Raleigh NC
    t.taylor@idcon.com
    ------------------------------



  • 11.  RE: PM Backlog

    Posted 01-23-2020 10:35 AM
    Hi Derek,
    This looks like a great opportunity for you to influence the maintenance reliability mindset in that organization. A few suggestion without knowing all details or specifics of the problem at hand - You may want to get some experienced frontline workers in confidence and make the PMs relevant so they have more ownership and accountability for it alongwith their supervisors. If the PMs are not done (thats why they end up on backlog)  but the plant keeps running with reactive maintenance , to me , this is a cultural problem.  We had good success in a previous role I had in mining  by really giving the frontline mechanics and electricians opporutnity to participate in planning process to eliminate non-value added activities, reduce holiday call outs and even reduce the PM downtime as people and process started becoming more efficient. We used the SAP reporting through business objects and made them visible at all levels from supervisors to GM. This was in a processing plant with lot of slurry pumping, complex piping and valving and complicated control systems which you probably have in the petrochem plant as well.
    Once that is done, you can start implementing RCM based approach for condition based monitoring etc. Be sure to do trade off analysis for deciding technologies to implement as some condition monitoring technologies also end up needing basic maintenance (for example automated vibration monitoring systems we had needed weekly checks to ensure sensors are in position and are reading correctly- comes with a cost).
    Also, I mentioned in the first paragraph about involving some frontline employees in the planning process - this is a good segway into trying Agile methodology for ensuring work actually gets done! Feel free to reach out if I can be of further help!

    ------------------------------
    Akshay Kulkarni
    Tucson AZ
    ------------------------------



  • 12.  RE: PM Backlog

    Posted 01-24-2020 01:28 AM
    Derek, 

    The problem you have mentioned is observed in each and every plant.  The details that you have mentioned do not make me fully comprehend with the operating context of the system. So, i would suggest the practical solution how i overcame , suggested and consulted . 

    1. Quality of PM: You can measure quality of PM  by seeing the number of corrective maintenance or emergency work orders between two PM. If there is corrective maintenance between two PM (PM on x date) and next PM on y date then there is an issue with effectiveness of PM . You may need to squeeze the interval or observe the maintenance practice and quality of maintenance.
     You can also measure it by Time to do the maintenance and trend it and check the difference between trends of time of maintenance.

    2.  Through your post, it seems that management is not ready to go for PM because it might have been observed that PM did not pay organization in the longer run and still they had failures after doing PM so it may be a matter of trust . It would require a full RCM based approach . Identify the system , put the system boundaries and perform full fledged RCM and come up with tasks of non intrusive maintenance like Vibration analysis, oil analysis, IR etc. 

    3. I also see that there is no swapping between the equipments which means that your two same equipments donot have running to standby ratio to overcome same age failure . It should be 70:30 ratio or 60 :40 . Now, probably there may be some operating context or culture which i am not be aware but it seems it may have happened last time they stopped one equipment and when other equipment was called on demand it didnot work. Hence its a hidden failure. So, when you perform RCM it would project through failure modes that perform function testing of standby equipment.

    So, crux, first go for a full fledged RCM program on system level.  Identify the failure modes, effects and consequences through risk based approach and optimize the frequency of tasks and make it a living program and check the quality of PM if ts a time directed task.



    ------------------------------
    Sameer Tiku
    Senior Staff Customer Reliability Engineer
    GE Digital
    Bengaluru
    ------------------------------



  • 13.  RE: PM Backlog

    Posted 01-26-2020 01:57 PM
    Since the facility is over 20 years old and you are new at the facility, You' better make sure that the contractor companies are good in their field. In the first year, when I faced a similar situation, agreed with contractor companies, and in the following year, shared the works for my team by following the steps suggested by Bruce Hawkins. So I gained a year extra time.

    The point to be considered here is the presence of spare parts. If an accurate monitoring system is set up, you can perform planned repairs at the first stage so that you are one step away from reactive maintenance. 
    If the contractor companies consist of the authorized services of the equipments, you may ask them for advice on critical spare parts.


    ------------------------------
    Serkan Divilioglu
    TPM Manager
    pharma
    ISTANBUL
    ------------------------------



  • 14.  RE: PM Backlog

    Posted 01-27-2020 08:25 AM
    Lots of good information in the comments and all are warranted ideas.  It will truly help you to know the causes of the failures and try to eliminate them.  The PM program will help to isolate it some by reviewing your times between PMs and the number of failures found in that time frame.  This will also help optimizing your program to efficiency.  Look into the current failure situation and define what it is.  There may be a simple fix to cut down the number of reactive items that you have and gauge how to proceed with Bruce's plan of PM implementation.  Hope that is helpful

    ------------------------------
    Michael Guns, Jr., CRL
    Associate Director, Maintenance Strategy and Planning
    Facilities - Buildings Maintenance & Operations
    200 Academy Street - Newark, DE 19716
    Maintenance Center - Room 113
    Office: 302.831.4055 Cell: 302.358.9381
    ------------------------------



  • 15.  RE: PM Backlog

    Posted 01-27-2020 09:11 AM
    Great inputs by other participants...to add my thoughts on this...many of them would be repeat also...
    1. One-time review and execution of PMs in PM backlog in a Shutdown opportunity would be effective only if these PMs have a clear basis. Else these invasive PMs might add to the Maintenance Cost and also introduce new defects into the Assets...The objective would be to do the right PM at the right interval and at the right time vis a vis clearing all the PM Backlog...

    In my view, the issues mentioned of high Reactive Maintenance, Low compliance to Equipment Changeovers, Low compliance to PM Schedules need to be looked into Systemic Improvements rather than one time actions to clear the backlog. You'd need to focus on Planning & Scheduling, Change-Overs, Root Cause Analysis, Condition Monitoring and Reliability Centered Maintenance...

    Focusing and working on the following aspects, based on strategic plan, would help:
    - Maintenance Work Planning and Scheduling (right maintenance work at the right time by the right people)
    - Effective Equipment Change Over with required compliance (including Procedures for safe and smoot changeover where the system is designed, alignment of the Change Over Schedules with the PM Schedules, training the field operators to perform the Changeovers as a routine)
    - Quality of field work execution and feedback on work execution/work history, implementing Precision Maintenance
    Good Condition Monitoring Program
    - Effective Root Cause Analysis for the Failures feeding back to the various systems/practices; Bad Actors identification and their management
    - Review of PM Program for effectiveness (feedback of PM work execution, whether the tasks address the right failure mode/s and their optimum frequency...)
    - Reliability Centered Maintenance based on RCM/FMEA and the development of Asset Strategies that are risk-based leading to an effective PM Program

    Hope this helps, thanks

    ------------------------------
    Sundar Naranammalpuram P.
    Navi Mumbai
    Maharashtra
    India
    ------------------------------



  • 16.  RE: PM Backlog

    Posted 01-29-2020 08:34 AM
    Excellent question and response from community friends. I would like to just add one point:

    LEADERSHIP 
    Do they understand the value of PM and impact of long term back log or CM to bottom line $ ? It appears not as they might not have the data in front of them to revisit their paradigm shift around PMs.

    ROI (Return On Investment)
    Can you develop a ROI program to justify the PM programs being done timely Vs cost of unscheduled repairs and justify.

    hope this helps,

    ------------------------------
    Krishnan Shrikanth
    CMRP
    SINGAPORE
    ------------------------------



  • 17.  RE: PM Backlog

    Posted 03-05-2020 03:27 PM
    Hi Derek,
    First step is to define and separate out "overdues" from "backlog". 
    Backlog = amount of work you know you have to do but haven't done yet.  Overdue = not done by the target completion date.  Not the same thing.​
    When you start calling overdues "overdues" then you tend to drive better behaviors.  Calling something out for what it is, is a good way to change behavior.

    Cheers,
    Mike

    ------------------------------
    Michael Spence
    Senior Reliability Engineer
    Refining NZ
    Marsden Pt
    ------------------------------