John Reeve
The link
that you provided on the

quote:

The Elusive Weekly Schedule.

does not open.
May you attend to providing another link, if possible.

Otherwise, the discussion has been live and informative.

Ecky

Very interesting discussions, indeed. To share some of the team's experience in Saudi Arabia:

1. ISO 14224:1999(E) defines corrective/preventive maintenance as:

Corrective maintenance: maintenance carried out after fault recognition and intended to put an item into a state in which it can perform a required function [IEC 60050-191:1990]

Preventive maintenance: maintenance carried out at predetermined intervals or according to prescribed criteria, and intended to reduce the probability of failure or the degradation of the functioning of an item [IEC 60050-191:1990]

Fault: state of an item characterized by inability to perform a required function, excluding such inability during preventive maintenance or other planned actions, or due to lack of external resources [IEC 60050-191:1990]

2. We have been using CMMS for over 8 years in several steam/gas power plants in Saudi Arabia. We have a simple-to-follow classification system. Those orders created by system (i.e., CMMS) are classified as preventive maintenance (PM) orders and those created by maintenance planner after fault recognition (i.e., ‘defect noted’ step) are classified as corrective maintenance (CM) orders.

Note that:
a) PM orders can be either be automatically triggered (based on time-intervals) or manually triggered (based on time-intervals/equivalent operating hours and annual planned outage schedule). A pre-requisite for PM is that the PM packages should have been approved by Technical Support Division Engineers and should have been configured in CMMS. Once configured, PM orders do not require planning efforts. Scheduling efforts will be, however, needed before PM orders are printed. Scheduling is done after confirming the availability of required resources (like spare parts, maintenance manpower, etc.). One more advantage of PM orders is that zero-based maintenance budgeting is possible, as level of PM work including quantity of component requirements are already known

b) Few power plants historically classified the Major Preventive Maintenance/Testing & Inspections (T&I) done during annual Planned Outages as PM, while other plants classified these as T&I. We are attempting to unify this!

c) Fault is recognized normally by Operations personnel, who in turn create Work Requests (Work Type: CM) in CMMS. In special cases, authorized Maintenance/Technical Support Division personnel can also create such Work Requests. Maintenance Planner creates CM order based on these Work Requests. CM orders need planning as well as scheduling efforts. One disadvantage of CM orders is that zero-based maintenance budgeting is not possible as the level of CM work are not known; only history-based budgeting is done

d) Any corrective work arising out of PM work, which is outside the PM order scope (example: additional spare parts needed, requires more than the duration indicated in the PM order, plan task for another crew) will be communicated but will be addressed through a follow-up order. Since the follow-up order will be created by maintenance planner, it will be of CM type

e) Condition Based Monitoring (CBM) data collection is done by specialized crew in some power plants. In other plants, operators have been trained specially to collect CBM data. Corrective actions are recommended through CBM Reports and are addressed through CM work orders. Though CBM work type is available in CMMS, there was a notion that this type should be used only when CMMS created orders automatically based on CBM data uploads. We think this notion is incorrect and needs to be corrected

3. There was a technical problem when we had upgraded CMMS in 2006 with regard to history and we trying to resolve this through the vendor. The latest version is running from June 2006 and the 26 months statistics reveal that PM (53.6%), T&I (4.4%) and CM (40.8%) constitute nearly 98.5% of orders created (103,093). Remaining orders are of following types:
• Special projects/modifications
• Commissioning (new equipment)
• Non-Equipment related
• Condition Based Maintenance

Note: Edited on 6 Sept to a) delete Site CMMS Coordinators in point-2(a) b)To introduce line space

This message has been edited. Last edited by: ganesh, v.j,

Dear Ganesh

It is very good that you are willing to share your experience with us here. Just a few comments to your points:

a) It appears that you attach bills of materials to your PM works such as overhaul of rotating equipment.

b) Planned outages during shutdown or turnaround should be considered PM regardless of when these are done. We can distinguish between PM during normal maintenance and PM during shutdown in the PM Work Type level if you wish for analysis purposes.

c) What work type do you use to raise unplanned/unintentional breakdowns and planned/intentional breakdowns (or run to failure)? Do you lump them together under CM? Is it nececassry to distinguish them?

d) I think there should be another work type i.e. CP for Corrective/Preventive to cater for all follow up works founds during PM and also from CBM. This is to show how effective your PM and CBM in detecting defects or abnormal conditions.

e) We define PPM (Planned Preventive Maintenance) to comprise fixed time maintenance, predictive maintenance (or CBM in your case), inspection and testing works. Again we can distinguish them at the PM work type field. This definition jives with the second ISO definition above i.e. according to predetermined intervals or prescribed criteria. So a separate work type for CBM is not quite necessary.

We have a plant change (PC) work type for project works including modifications and commisioning which is part of project execution. Do you track temporary plant changes under this work category?

We also have a non-core maintenance for miscellaneous and standing instruction jobs.

Wouldn't it be nice if industry (we) could standardize on worktypes, whether we use them all or not? Better yet, if we adheared to standardized worktypes, our data could be compared across the industries without translations or caveats.

Dear Wally,

I agree with you with regarding to standardizing work types.

We came across very interesting local as well as international references including SMRP, ERPI (‘Optimizing Power Plant Maintenance’ article by Patrick Abbott in POWER magazine, December 2004 issue), few ISO standards, this discussion forum, Saudi ARAMCO’s General Instruction on Maintenance Work Orders, SABIC/UNITED SAP-Plant Maintenance Work Order Types, etc. But there seems to be no standard in classifying maintenance work types across industries!

We have over 40 power plants distributed throughout Saudi Arabia; some of these are power and desalination plants. Our team here has been trying to unify the plant maintenance and project system business processes for our company’s generation power plants nearly for a year. One item is standardizing plant maintenance and project systems work types; standardizing work types is only half-complete!

Regards.

------------------------------------
Dear Josh,

Appreciate your detailed feedback.

Point-wise response to your comments:

a) The comment is not clear to me. We do T&I for rotating as well as stationary equipment like heat exchanger, condenser, steam generator, etc. T&I is for main equipment, their auxiliaries and balance-of-plant. Note that the terms ‘Shutdown Maintenance during Planned Outages’, ‘Testing & Inspection’, ‘Turnaround’, ‘Annual Maintenance’, ‘Annual Overhaul’, etc. refers to ‘T&I’.

b) I personally agree with you that maintenance done during planned outages (aka T&I) is also a PM. Two of big power plants in our company had been historically classifying this as PM. But, few others were classifying this as a separate work-type (T&I) and agreed recently to change to PM in principle, that too after a heated debate! T&I require long-lead materials, capital spares and other special resources like T&I contractors, OEM experts, huge cranes, etc. These resources need to be planned or/and budgeted well in advance. Moreover, T&Is are costly and complex and is done in a similar way to a project. Hence, we cannot deny there is a practical advantage of classifying T&I as a separate work type, as this will help us in planning, budgeting and controlling besides analyzing.

c) Unintentional breakdowns were classified as CMs with ‘A’ or ‘Very High’ or ‘Emergency’ priority. Some attempted to classify this as ‘Emergency Maintenance’ in the past but discontinued later. This is because regardless of the urgency, it is a CM job. We do not have a classification for planned/intentional breakdowns yet, but we have a proposal to introduce ‘RTF’ (Run-to-failure) as a sub-type of CM

d) We have a proposal to address this using a sub-type of CM called ‘corrective work identified from PM/CBM’. We are still debating on this. One school of thought is any corrective work identified from PM/CBM should also be classified as PM/CBM and not CM! How about that?

e) Please see the response for item-b above. Instead of ‘fixed-time’, it would be better to use ‘fixed-interval’. PM or T&I for some of the equipment are not based on ‘fixed-time’ but on ‘fixed-counter’ intervals like EOH (Equivalent Operating Hours). We think the definitions cited (ISO 14224) are quite practical.

f) ‘Plant changes’ is a good terminology. We thought of classifying this under ‘Improvements/Modifications/Upgrade’ work type instead of currently used ‘special projects/modification’ CMMS work type. Possible sub-types include: Capital modifications/upgrade project, non-capital modification/upgrades, engineering studies, work arising out of continuous improvement studies, proactive work as a result of energy audits or environmental studies, etc. What are temporary plant changes? Can you please give an example? Does this mean bypassing some interlocks due to, say, defective instrumentation? If so, this is handled by multi-task CM work orders.

g) We will probably do away with separate ‘commissioning (new equipment)’ work type as this was primarily intended to track warranty aspects of new/upgrade projects. ‘Warranty’ sub-type should do

h) I think ‘Non-core maintenance’ is a better term than ‘non-equipment related’ work type. Will discuss with our team members here. What is standing work instructions jobs? We have something called ‘standing work orders’, but this is not classified as a ‘work type’.


Regards.

a) Do you attach spare parts for PM to the PM plan itself? I think yes bcause of your statement here "...quantity of component requirements are already known". However, what do you mean by "zero-based maintenance budgetting"? Does T&I stand for Turnaround and Inspection?

b)If you wish to track T&I work orders, we can use another PM work type or use the work centre or the separate project budget code in the cost centre.

c) It's good that you want to introduce RTF. One time I mentioned about this RTF but the first response was " Do we want to be as complicated as that?".

d) I agree that corrective/preventive work from PM should be classified under PM. And presonnaly, I would like to see what would the percentage in reality.

e) Yes fixed interval is more emcompassing term for fixed calender time and fixed counter time such running hours.

f) I think your definition of plant changes is more comprehensive because it includes engineering studies, continuos improvement, energy/envirinmental studies etc. Yes, temporary changes include bypassing interlocks (the most important one actually to mange changes in DCS, BMS, PLC etc). Other that, some piping repairs can be in this category. The UKOOA integrity management document does list these items clearly.

g) I think SAP PM can track warranty for neq/upgrade equipment if confugured so but I haven't used this feature yet.

h) I guess standing work instructions and standing work orders are the same thing but slightly diferent terminologies.

Dear Josh,

Appreciate your prompt feedback and thanks for clarifying some of my doubts.

Point-wise replies:

a) Yes, we do attach required spare parts/materials to the PM plan itself. Not all PM plans need parts/materials. For some PM plans, the required materials will be withdrawn in practical quantities from warehouse and stored near the work shops (example: lubrication oil/grease).

Zero Based Budgeting (ZBB) is an approach to budgeting that starts from the premise that no costs or activities should be factored into the plans for the coming budget period, just because they figured in the costs or activities for the current or previous periods. Rather, everything that is to be included in the budget must be considered and justified. In traditional incremental budgeting (i.e., history-based budgeting), department/division managers justified only increases/decreases over the previous year budget and the base (that was actually spent during the previous year) is justified by default!

T&I is ‘Testing & Inspection’ (See 27 Aug 2008 posting, item-2b)

b) Some power plants were initially insisting that their use of separate work type for T&I was correct. Some power plants were using PM work type regardless of whether it is carried out during planned outage or not. Proposal to change the existing practices to T&I-PM, Non-T&I PM, PdM/CBM has been agreed in principle. We are yet to finalize whether to put these as sub-types of PM or as separate work types. We are aware that CMMS/Avantis v4.1 supports many work types. We are not sure about how SAP will handle work types & sub-types. What is the relationship between SAP-PM order types and Maintenance Activity types?

Your suggestion to track T&I work orders using separate ‘project budget code in the cost centre’ is a good one. Please see the practice followed by a reputed Saudi company (which has already implemented SAP-PM):

“.. a budget is created against an Internal Statistical Order (Type ZITI), representing the total budget for the T&I by budget year, and approved by the proponent. A ‘Main’ PM03 Work Order is created (via an approved M5 Notification) referencing to the Internal Order. To maintain the Internal Order link, any subsequent Orders for the T&I are created as ‘Sub Orders’ of the ‘Main’ Order. The total actual cost of all associated Work Orders of type PM03 are incurred against the Internal Order as settlement occurs…”

How you are able to track T&I work orders using work centers? Does this imply that T&I will be always performed by contractor? Or, does this imply specific company work center(s) for T&I? In our plants, T&I jobs are done partly by contractor and by employee groups/crews. Some employees carry out CM/PM work order tasks for running units, while most of the employees from the same groups/crews carry out T&I/PM/Deferred CM maintenance tasks of units under shutdown.

c) It is true that during unification meetings here, some of our team members asked the same question with regard to RTF. We historically did not classify RTF as a separate type; it was lumped into CM. The proposal to introduce it as a sub-type of CM is now gaining momentum. It is a step in the right direction.

d) If you apply the concept that ‘corrective maintenance is maintenance carried out after fault recognition’, the follow-up work order to correct the fault discovered during PM/PdM/CBM/T&I activity shall be classified as CM. If you apply the concept that ‘preventive maintenance activity is intended to reduce the probability of failure or the degradation of the functioning of an item’, the follow-up order shall be a PM. There were interesting references in the forum and the debate is still alive! The debate is compounded by some practical problems such as:

(i) How to classify ‘faults’ and ‘incipient faults’ objectively?

(ii) How about CBM/PdM data collection work done by specialized crew (like vibration monitoring group, motor condition monitoring group), etc? A proposal was made here to trigger ‘CBM/PdM data collection’ work orders using the Preventive Maintenance Master Data. All liked the proposal and agreed to implement it.

(iii) How about plants that have trained operators to collect CBM data? Should we trigger orders? We seem to dislike the very idea of triggering orders for operators. We rejected it.

(iv) Operators are performing pro-active tasks periodically like valve-freedom tests, starting of emergency-diesel generators/jockey pumps, change-over to standby auxiliaries, etc. These pro-active tasks are done conventionally without work orders! Should we start accounting these by triggering orders? FYI, one large steam power plant had even a data base and named it as ORAMS, Operations Routine Activity Management System. Others use simpler ‘daily/weekly shift items’, or use ‘shift/operations log’.

To resolve this, our team:

(v) Introduced a ‘defect noted’ (same as ‘fault detected’) as the first step in the ‘Unified CM business process’ flowchart and all agreed. Regardless of the type of faults (functional fault, breakdown, incipient fault), CM work order needs to be created. Based on this logic, follow-up work orders from PM/PdM/CBM/T&I shall be classified as CM. One proposal to address these ‘discovery’ work orders using a CM sub-type is currently under discussions. At least one sector consisting of 15 power plants has agreed to the proposal and two more sectors raised no objections.

(vi) We agreed that the preventive maintenance master data (aka PM plans) shall be reviewed, approved and configured in the ERP system before PM Orders are triggered. This is the practices in the plants using CMMS for several years. It has been proposed that sub-types like PdM, etc shall be automatically generated by system based on configured preventive maintenance master data.

With regard to percentage, consider the following theory:

“When establishing an effective maintenance program, one must determine not only which Preventive Maintenance (PM) routines to accomplish, but how often should they be done. The answer to this question would seem on the surface to be quite simple and, in fact, one proven theory is that the PM to Corrective Maintenance (CM) work order ratio should be about 6 to 1. This theory assumes that the PM inspections should reveal some type of corrective work that should be completed on an asset on average every 6 times it is accomplished. The assumption is that, if the ratio is greater than 6:1 you are performing the PM too often; if the ratio is less then 6:1, you are not performing it often enough. (The “6 to 1 Rule”, proven by John Day, Jr., Manager of Engineering and Maintenance at Alumax of South Carolina, during the period when Alumax of South Carolina was certified as the first “World-Class” maintenance organization).


e) Thanks. We are marching together towards ‘standardization’, this should make Wally happy! At least some people started listening…

f) Historically, many managers disliked the use of ‘interlock bypass register’ kept in the control room and insisted on using of CM work orders.

I need to go through ‘UKOOA integrity management’ document.

g) Our understanding from one company, which is using SAP-PM here, is that they are using PM activity type M11: Warranty Work in work order

h) Thanks.

I have been avoiding to join the forum, though I was getting benefit out of it. Our project coordinator and team-lead only had advised me to write.

Regards.

Ganesh, you write more than me. It takes time to read and understand. Anyway, some studies states we understand the subject better if we explain to others. So your decision to participate actively in this forum is positive.

a) Ok, now I understand that ZBB now.


b) In SAP PM, each work order type can have many Maintenance activity types as you configure(ie. one to many mapping). Pls note some CMMS do not have the Activity type so they have many work types.

The work centre should be assigned to internal staff group who manages and supervises the work execution by the contractor.

c) Ok

d) Yes, we can separate the follow up work orders into corrective or preventive. It's somewhere in between. That's why we use Corrective/Preventive (CP) work type. However, we can opt not to do anything for the corrective work so any action taken can be considered preventive in nature.

i) Why do you need to differentiate between fault and incipient fault?

ii)Yes WO should be generated for any works done rgdless of who do it.

iii) Yes, WO should be generated for works done by operators. At the end of day, we want to se the percentage of works done by operators with some organizations advocate operator-drive maintenance and reliability. This appears to be out of your comfort zone but operators already raised notifcations in our case so processing WO is just another additional step for them.

iv) Yes, once operators are climatized to WO, it's easy for them. My stand is all plant works should be managed using the CMMS for one-stop data centre for plant KPI analysis.

Having another database is not new to me because some companies manage shutdown/turnaround works separately from normal maintenance works, using the so-called specially customized software but indeed they don't really see the fully-integrated work management system and still thinking in a compartmentalized way! And the KPI analyst will have difficult time to integrate and reconcile both data banks in order to have a comprehensive analysis!

v) What is discovery WO? Is it works found after opening equipment eg. during shutdown? If yes, we call this emerging works which should be treated as corrective/preventive works if you wish.

vi) Any basis for the 6 to 1 rule? Does it mean for every 6 PMs, we should have 1 follow up work order? Sound good but is achievable?

f) Yes, we should record interlock bypass register in CMMS plus all other changes to plant softwares.

g)Yes, PM Activity Type is a subset of PM Work Order type. I notice that the plant using SAP PM uses PM03 and M11 which are standard in SAP PM and not using the direct abbreviations of work types such as CM, PM, BD, etc. It can be quite difficult to memorize for users although with definitions inside the system.

Dear Josh,

Thanks for your prompt response. Will try to be brief this time!


b) Thanks for clearing one-to-many mapping between Maintenance Activity Type and Order type. We will discuss on work-centers separately later.

d) Your use of CP (Corrective/Preventive) code is good; some were suggesting equally good ones such as ‘PM-Repair’, ‘PMO’ (PM originated work) and ‘PMC’. These are all in line with Leon’s suggestion (11 March 2008 posting by Terry) to capture follow-up maintenance in a separate work order and code them as such. SMPR is also suggesting something similar.

If you are using CP type, what is the ratio of CP to total PMs? Leon says follow-up maintenance work from PMs should amount to 12% to 20% in a lean maintenance environment. This is also in line with 6:1 rule of John Day, which works out to 16.7%.

While going through all these forum discussions again, I got some more ideas for the codes: FM (Follow-up maintenance to PM or PdM or Testing & Inspections, etc.), DW (Discovery Work).

You did suggest a rule of thumb with regard to classifying minor follow-up maintenance work as PM by adding operations/tasks under the same work order. Are you still following it in your organization? Will it not make tracking of costs difficult?

(i) Mike used the concept of going beyond the ‘incipiency’ condition to differentiate between BM & CM - that is the reason why I asked.

Note that we do not subscribe to the idea that corrective work orders following a breakdown are not scheduled. All CM work orders other than the ones with ‘1’ or ‘A’ priority are scheduled! Some less critical equipment that are intentionally run-to-failure (like circulating water pump traveling water screen motors). Even these breakdowns are currently classified as CM. RTF sub-type needs to be approved.

(iii) Our team here has just agreed to discontinue with ‘minor’ type of work requests for minor maintenance work carried out by operators (like oil-filter replacement) without work order. One reason: cost will not be captured in work order. Now you are suggesting to create ‘work orders’ for operators. Though your proposal is in line with modern maintenance practices, I think this will be a bitter pill to swallow. ‘Change management’ efforts required will be very high!

Are you practicing this in your organization? Since when?

(iv) Same as above

(v) You got it. Discovery work is same as CP.

Regards.

Sound like you are analyzing all the posts in this forum. Is it for a paper or consultancy?

We only add some minor follow up works into the same work order if less than 2 hours to complete. Just do it on the spot!

We generate work orders for operator basic care tasks or functional test by operators but be cunning in managing the seemingly many work orders with small manhours in each.

For change mgmt, identify someone who is good at CMMS to champion the cause to start the ball rolling.

This message has been edited. Last edited by: Josh,

ganesh - whatever your system it is only going to be as good as those that use it.
I know the catch phrase KISS applies to our site esp. when all tradesmen are responsible for originating PMs - that is why we only stick to the common ones:
SF = service fitting
CR = corrective repair
BR - breakdown repair
CP = capital project
PM = preventive maintenance routine (scheduled)
Even then they tend to get it wrong either accidentally or don't care.
I guess if a maintenance planner inputs all your data though it is different.
Comes back to what I said earlier. Decide on a system and stick with it (get user buy in first though).

Mike.

We are scratching deeper here, aren't we?

Dear Josh,

Yes, we are analyzing all the posts in this forum. Not for a paper/consultancy. We are unifying the maintenance business processes (flowcharts) here, in Saudi Arabia.

Sure, it makes practical sense to hook the follow-up maintenance in the same order as it takes less time than to create, plan and schedule a new one. If actual costs of the follow-up are much more than planned ones, it is a disadvantage; this is because PMs can be (and are) usually planned accurately (as against T&I sub-type of PM). Some of our managers here think PM work orders are part of approved PM program (i.e. system created and not planner created). I personally agree to this view as it is easy to practice.

We noticed similar problem here: for small maintenance work like cleaning/replacing filters, operations created more 'minor' type notifications for themselves! One power plant even considered this in their operation-groups’ productivity calculations.

We do have a CMMS champion in our group. Additionally through discussions, mails and meetings, we have started the change management already.

Please go through the attached presentation – I am not sure it is permitted or not. I am taking the risk.

Regards.

Dear Mike,

Here, operations will indicate the work type in Work Request in CMMS but planner can change it while creating the order.

I agree with you: garbage in, garbage out applies to any system. It is ultimately the technician who need to enter closing comments properly instead of entering just 'PM done' 'job completed'!

We have been discussing your idea: decide on a system and stick to it! Since we have an opportunity to move to a new system, discussions are on to decide the changes.

I will apply for a KISS waiver

Regards.

This message has been edited. Last edited by: ganesh, v.j,

The attachment please.

PLANT_MAINTENANCE_WORK_TYPES__WORK_PRIORITIES_v4.2.ppt (848 Kb, 32 downloads)

VJG,
Thank you for sharing your work, which is both thorough and good.
I have some suggestions on Priority definitions, using the following bullets:

- If we accept the purpose of maintenance as one to reduce the risk of failure (Prob. x Cons.), High Risk equates to High Priority
- Both PM and CM can be of High Priority, if they relate to downtime of critical eqpt. The more the proportion of PM in this list, the more in control of the situation we are, and hence a desirable goal. In this context, the use of the word 'Emergency' implies an unplanned event, not a PM activity. Hence I do not support its use.
- Preset Priorities may need upgrading, if e.g., the reliability of Standby eqpt. deteriorates while Duty item is being repaired.
- Urgent work is NOT the same as High Priority work; , i.e, if the risk level was low/medium, but the urgency was due to purely admin reasons (e.g., initiator forgot to plan/schedule on time, delayed availability of spares/logistics etc.). We should not equate Urgent with High Risk, though it is often (but not always) true. For example, preparing the site for a VIP visit can be Urgent, but it has no impact on the Risks to Safety or Production - the only risks may be to your job!
- The Aramco idea of tying the Priority approvals to Authority levels is good.

Our goal must be to focus of the risks. Hence the work priorities must match these risks. Eliminate every admin. urgency as far as practicable. Use SAP to capture such inefficiencies.

Ganesh

Thanks for sharing the presentation.

All PMs should be approved and planner can only create them after the approval. Most PMs are created upfront in once-off exercise but be prepared for the subsequent changes as necessary during and after implementation. There should a procedure and focal point in place to manage these changes.

Vee, I agree with your points.

The idea of upgradable priorities for eqpt with redundancy is known but I think it's done by operators when raising the work request. In fact, the equipment criticality that actually increases when one of the equipment not working.

The risk (=prob x Con) ranking can be done in CMMS if RBI exercise is done first.

People are often confused by high priority and emerngency or urgency. Tough to explain.

hmm and for some people every w/o they input is an emergency!

That's why we should have proper planning and scheduling in place.

Mike,
I think you will find Chapters 25, 26 and 27 in my book '100 Years' interesting in this context.