Join or Manage Your Profile
Posting Boards
Maintenance and Reliability
Posts About Improving Reliability
Retention period for machine history|
Go
![]() |
New
![]() |
Find
![]() |
Notify
![]() |
Tools
![]() |
Reply
![]() |
|
Our corporate document management people have decided that maintenance records are to be kept for five years, then destroyed. (They are not maintenance people.) As a reliability professional, I think this will be a serious mistake, to begin destroying maintenance history over five years old. I would think that machine history should be kept for at least the life of the machine, and even for the life of similar machines.
Are there any published guidelines or standards addressing this, that I could present to our document management people to back me up when I ask them to change that retention policy? I'm pretty sure they won't do it on my word alone. This message has been edited. Last edited by: Mike Dayton, |
|||
|
Sadly I do not know of guidelines or standards to help you. In my company a few years ago we got a similar corporate policy on document retention.
But you still have data eletronically stored at the CMMS, right? When you close the Maintenance Orders, do you type in whatever comment the technician wrote down on the order hardcopy? Darth Eugene Vader |
||||
|
Our maintenance technicians type their comments directly into the CMMS, and when they close work orders it becomes part of the electronic machine history.
This issue came up this morning because I overheard our local "document sheriff" asking questions about the CMMS, and she mentioned the five year retention. |
||||
|
These 5 year periods apply mostly to non-technical (financial) paperwork etc..
In some banks and other institutions this is a legal issue, because they are getting/supplying tons of paper every day. Imagine you have 50.000 clients and you have to generate paper(work) for a lot of transactions like notifications of address change, payment of debts etc.. Lets say you have 500 pieces of equipment (assets), I can assure you that you have far more less paperwork to take care of. In my opinion destroy the paperwork when the machine (asset) does not exist anymore. If you sell it, hand all the documentation (parts + manuals) to the new owner. 1) One should start with asking the reasoning behind destroying maintenance records. 2) If the answer is "corporate policy" or the ISO manual said so, my next question would be: what do I gain with it? If you are in a nuclear/chemical plant and you destroyed your evidence of mandatory inspections because of some arbitary chosen time, expect to suffer a lot of headache Steven van Els, CMRP |
||||
|
Another issue:
If your CMMS changes next year, will you be able to transfer your history electronically to the new system??? Steven van Els, CMRP |
||||
|
I agree with you, Steven, that we should keep these records as long as we own the machine.
As for the CMMS, a bit of background. I work for the North American division of a worldwide company. A couple years ago, there was an effort at the corporate level to have all North American facilities begin to use the same commercial CMMS. It was to be a three year project, so for the interim, we developed a home grown, web based CMMS. Well, the corporate project fell through, so now we have our interim solution for the indefinite future. Given all this, we are better off than before, anyway. And since the home grown system was to be only interim, portability of the data was a primary requirement. What concerns me is that we have an electronic document management software that will automatically delete any document that is beyond its retention period. That software has not as yet been applied to the CMMS database, but it will affect other documents that we have, such as reports from periodic PdM inspections (in house and contract), etc. I am also concerned that we are in a position where non-technical people are telling us when we should destroy maintenance records. This message has been edited. Last edited by: Mike Dayton, |
||||
|
I agree that the records should be kept for the life of the equipment type. An occasion in recent times to support this is when a court case required the hard copies of work orders to be produced as part of evidence for a compensation case. Hence the company were unable to produce these with the technicians signature on them. This in turn created a huge headache for the company.
Apart from the legal implications I agree with what was mentioned above regarding the changing of CMMS packages, I have seen a number of occasions where the software has been changed and the history not able to be or just not transferred as it was not part of the contract. On occasions to retrieve this history from the old system was quite an expensive exercise to get converted to a usable format. |
||||
|
I had to re-read this several times before accepting the premise, and only hope your view wins out in the end. As a case in point, just last year I found myself digging up a couple of files created using PC-Outline during the mid-80s in order to diagnose and repair one of our machines. When we get new equipment it isn't uncommon to go through a 2 to 5 year learning/debugging cycle until we've a good grasp on what things are liable to break and how to efficiently repair them, and have done minor redesign on weak spots along the way in order to decrease failure rates. In my opinion, records created throughout this period remain useful, and often prove indispensable. Unless those components which are part of an on-going condition monitoring record system fail after every five years or less then why even bother to look? Often, such records become more useful over time; not less. Seems to me throwing away this hard-won knowledge after an arbitrary 5 year period, and for no particularly good reason is insane. |
||||
|
Mike and others,
Why 5 years? Why not 3 or even 1 year? How did this magic number appear? Let us look at some technical arguments. 1. Machinery components degrade over varying periods, Some elements, such as clutch plates may do so in months, while others like welds may take 5-20 years to do so. Pipelines may corrode/erode over a 10-30 year life. Concrete degradation can take 15-30 years. To track degradation, we must keep records at least as long as their normal degradation period. Else we won't know when to replace them. That is a Capital Investment decision, one that should be close to the bean counters. 2. Maintenance produces useful production capacity; that is a primary reason for doing it. To use this capacity effectively, we should be able to predict available capacity, so that Production Planners can optimize machinery use and hence profitability. To predict capacity, we need to analyze failure data. With good management, we can expect, on average, failures to occur not more than once in 2-5 years (or an MTBF of 2-5 years). Average means that sometimes it will be much less than this period, at other times multiples of the MTBF. To track and analyze failures, we need at least 5 failure points. In practice, some data points get censored or discarded, so we are talking about 7 or 8 data points. So for a single item, we may need 15 years of data. We can of course analyze families of equipment if they are independent and identical. In real life, these conditions are hard to fulfill, but we make approximations. Recall that all this is in aid to maximizing profitability, again something the bean counters are after. Next, some legal reasons; 3. If there is a fatality or serious environmental incident a la BP Texas, and the Regulators want to see our records, try telling them we have a policy of shredding all our records every 5 years! 4. State and National laws require us to retain data, including legacy data. This may vary in each country, but fall foul of these at your peril. A commercial reason: 5. In these days of M&A, when companies change hands fairly often, the potential buyers need to look at our data as part of 'due diligence'. Not having of inspection and maintenance records can seriously impair our commercial position, something our shareholders won't appreciate. Perhaps your sheriff is doing 'kite-flying' to test the water. Perhaps you can tell her to go fly a kite! Regards, V.Narayan (Vee) Lead Author, 100 Years of Maintenance: Practical Lessons from Three Lifetimes, Industrial Press.NY ISBN-13: 978-0831133238 Author, Effective Maintenance Management: Risk and Reliability Strategies for Optimizing Performance, 2004, Industrial Press NY ISBN-13: 978-0831131784 |
||||
|
| Previous Topic | Next Topic | powered by eve community |
| Please Wait. Your request is being processed... |
|

