Jump to content
Test ×

Derek Brown

  • Posts

  • Joined

  • Last visited

  • Days Won


Posts posted by Derek Brown

  1. Hi All,

    I am curious about the concept of random/age related failures and wondered...

    If 70% – 90% of failure modes are not age-related, how does this actually work in practice?

    On my plant we are getting a new item of equipment fitted to replace an old asset, although it is a like-for-like machine, (it is a rotating powder transfer valve - or RAL), does this mean that 70% – 90% of the failure modes on my new equipment are random and i should be utilising Condition Monitoring where i can?
    Also, what failure modes are time based only? is it the same failure modes all the time or can these change with asset/conditions/environment/design etc...?

    If we try and go with this above statistic, how do we use that on plant, conduct an FMECA to determine the failure modes, score them and then see how these modes are prevented?

    When I go to install my new asset I want my maintenance instructions to incorporate these probabilities of failures but how do I actually do this? - does this all link back to an RCM approach? if, so, we do not routinely do RCM/FMECA and we do not have anyone who is trained in RCM, as most of our maintenance procedures are the same since the plant was built 20 years ago, namely mostly time based maintenance. 




  2. Hi All,

    Just wanted to get your thoughts on Pm Backlog and your experience of them (if you have had such experiences)

    Currently on my plant we have a perpetual backlog of PM's typically because the plant was built on the cheap and little redundancy built-in, meaning that PM's have to be done during shut-downs, which we typically never get through. Management are also nervous about doing PM's on the run in case we trip the plant which is pretty frustrating.

    Just wondered if anyone else had this issue and if so, how did you manage to overcome it.



    • Like 1
  3. Hi Erik,


    Thanks for the reply.

    We do CBM but we most definitely utilise technology enough whereby we currently have a contractor who comes to the plant every 5 weeks and does their measurements and then leaves to issue a report 5 weeks later. Not ideal as we need continuous monitoring on a lot of equipment in my opinion.

    On another note @Erik Hupje, we have a large backlog of PM's we are now only doing (or trying to do) when we have an outage. The problem being that we never fully get through them all and where plant can be released on the run, the Ops/management are too risk averse to release anything for maintenance. That is the strategy they are using at the moment - extremely frustrating. We now have to start looking at using external technicians at outages to get through our PM's, to me, this is ludicrous. I have not had time yet to look at the backlog, but I suspect most if not all will be time based PM's.

    Have you had a situation like this where plant wont be released and the backlog is piling up and you cant get enough time at shut-downs to do them all?




  4. I was reading Ramesh Gulati, AEDC Best Practices Result in Reliability Book and in it he discusses how typically world class, or best in class plants have a ratio of 20:80 of time based to condition based PM’s. 

    I then thought about my own 20 year old petrochemical plant and I’d say we’re probably 20:80; but the other way round.

    It got me thinking, how do you get to that ratio? 

    How do you check if your PM routines add little to no value? 

    How do you actually make the switch from time based to CDM?

    We have a reliability team but they said until we have a CMMS that can be set up for CDM (Maximo 20 odd year old) we cant simply move over to condition based maintenance. Why not? I didn’t really understand how that can be when you simply generate a Work Order for a CDM maintenance routine.

    We routinely have a backlog of PM’s either because they are mostly time based and the plant will not be released to us, and we do not have a lot of redundancy assets.

    Anyway, if anyone has made the switch I’d love to hear how you did it, what was the challenges and any advice you have.





  5. Hi @Erik Hupje,


    Thanks for your response.

    I think it prudent to mention the number of start/stops will also have an effect on seal life, and with condition monitoring fitted, should negate the need for time base maintenance and prolong the service life between overhauls. The  strategy for the least number of starts operating in a duty/st-by arrangement I think would prove to be the best for lower maintenance and downtime.

    I suppose the two issues to consider are uptime and costs. 

    Alternate or duty/st-by policies both have their pro's and cons and in term of reliability for either depends on their operating context and age related failures would be evident for a machine that is running and a machine that is not.




  6. Hi there,

    I was looking for peoples opinion on the duty/st-by arrangement for pump sets. I have not long started at a large petrochemical plant and to my surprise they do not do a periodic swap over of pumps as the Ops team have their “favourites”.

    I am now looking to convince them otherwise and introduce an Ops routine in the CMMS to try and push them to do it. My question is has anyone ever encountered this and how did you overcome it, and also what was your strategy? For me having two pump sets with the same run hours is not ideal and should have a lead/lag machine where the lag is only used in times of inspection, overhauls or breakdowns until the lead unit can be brought back online.



  7. Hi @Erik Hupje,

    Many thanks for this.

    I already have the Reliability-Centered Maintenance by John Moubray, I bought it as my lecturer advised it was worth having.

    Re the certification, Reliability engineering is certainly the niche' engineering route I am keen to pursue further, although there are different interpretations of reliability, or asset engineer certainly in industry, where reliability can just mean the nuts and bolts and not data analytics, failure rate acquisition and root cause analysis. This is where I find myself at the moment as my role is predominantly "keeping the plant going" and not enough time to devise medium/long term reliability strategies. I know this has to be a management driven objective to change the culture of maintenance but it does take time and effort. I read a quote on reliability - "everyone thinks reliability is like safety, it's a great idea but not everyone does it" - Heinz Bloch. Although my employer is trying to build a reliability team at the moment, it is something I would like to be part of. I have been in my current role for a few months and I am keen to make changes where I can and improve the plant I look after and learn as much as I can from others and utilising technology.

    I have heard of CMRP certification and it is something I will likely pursue, but are there no UK based accreditations for maintenance and reliability?



    • Like 1
  8. Hi Eric,

    I am upstream, looking after a plant using propylene to make plastics.

    I studied Maintenance and reliability at university after being on the tools for a number of years and I was amazed how much I didn’t know. For instance, maintenance is taught as a science, statistics, machine failure modes, the physics of how machines operate and more importantly how they degrade and fail. So sites such as yours instantly peaked my interest.

    Are there any certification for reliability practitioners that you can recommend?



  • Create New...

Important Information

By using this site, you agree to our Terms of Use, Privacy Policy and use of We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue..