Dear all,

Within the frame of the EU Project called RIMAP (Risk-Based Inspection and Maintenance Procedure), project members are currently involved in benchmarking best practices related to the processing of maintenance backlog according to a risk-based scheme.

The main concern is that the planning of deferred maintenance is likely to collide with other scheduled daily routines thus needing some risk criteria for sorting out the short term performance of all this maintenance, eventually backlogging again other non-risky tasks.
We'd like to hear from you on any specific scheme you might have implemented in your plant (as per Corporate rules or plant practices) as well as the risk criteria you use in such case?

Best regards to all
Bernard

Dear Bernard,

I suggest you consider the following steps to help manage backlog.
1. Prioritize all maintenance work whether preventive, corrective or detective, according to its risk ranking.
2. Rank order the work by risk rank on X-axis and manhours required on Y-axis and draw a Pareto curve.
3. If your prioritization is OK, you should find that your work distribution is
- High priority 5-10% of total manhours
- Medium priority is 20-25% of manhours
- Low priority is 65-75% of manhours

If this is not the case, re-examine your risk ranking process.

Such prioritization must be updated periodically, usually weekly, to reflect the dynamic operational and technical integrity situation.

For time-based preventive work, high priority work should be done within a +/-10% band width centred on the scheduled date. Medium priority work can be done within a 25% band width and low priority work within 50%. Corrective work (those identified by condition monitoring, inspection, or as a result of a breakdown) will generally tend to be of high or medium priority, and should be scheduled accordingly.

Using these guidelines, and assuming that staffing levels are satisfactory, backlog can be managed in most situations.

V.Narayan.
Author of Effective Maintenance Management - Risk and Reliability Strategies for Optimizing Performance, Industrial Press Inc., ISBN 0-8311-3178-0

quote:

---

Originally posted by bvalette:
Dear all,

Within the frame of the EU Project called RIMAP (Risk-Based Inspection and Maintenance Procedure), project members are currently involved in benchmarking best practices related to the processing of maintenance backlog according to a risk-based scheme.

The main concern is that the planning of deferred maintenance is likely to collide with other scheduled daily routines thus needing some risk criteria for sorting out the short term performance of all this maintenance, eventually backlogging again other non-risky tasks.
We'd like to hear from you on any specific scheme you might have implemented in your plant (as per Corporate rules or plant practices) as well as the risk criteria you use in such case?

Best regards to all
Bernard

---

Dear bernard,
We started two months ago a pilot RBI project and we use criteria for fisk evaluation: financial losses,environmental losses, personnel illness/injury. Some more information you may find at www.tischuk.com.
regards
dimitar

i used one idea from old herbaty's book on maintenance management and it works very well in last 9-10 months:

-by definition, deferred job is job to fill blanks in scheduling

-we established independent priority system for deferred job (we have simple one with three levels of priority)

-scheduler allways fills in deferred job of higher "deferred" priority if estimated lenght of job fits the blank in schedule. if not, any deferred job that fits can be used to fill the blank

- what can easilly happen is that some lower priority deferred job never enters the schedule. periodic audits help here (we started with three months' audit). decision must be made regarding such job: priority must be changed or job can be given to outside contractor OR job can be dismissed. this must be assessed properly!

Dear Bernard,

I have had some great successes using risk matrices for works in progress that are encoded within the Work Order priority structure.

To build on what was mentioned by Vee:

Work order priorities are structured in such a way that they reflect a) the severity of the consequences if no action is taken and b)the time that can b e allowed to elapse before the probability of failure is "intolerable".

This sounds heavy but in practice it is pretty light and easy to direct. Whats more, it enables the work order prioritization system to become a logical one whereby there is a level of control over works in progress.

In the case of a detective task, for example. When it is created and instituted its frequency is determined based on a level of risk that we are willing to tolerate. (Detective tasks are for hidden functions normally speaking)

Therefore executing this task within as certain timeframe will enable us to manage our risk to within the tolerable level we started out with. If it slides past this date then we are outside of our tolerable risk levels and exposing ourselve to a level of risk that we have already deemed as being unnacceptable. (Similar could be done for corrective works in progress)

So, in answer to your question, I have had success with prioritization methods such as these, rigorously implemented, and reported on (controlled) via graphical reports such as "Priority versus time". Whereby you can see how many work orders, and potentially which work orders, have placed the organization outside of its acceptable risk envelope.

It also goes some of the way to general backlog management although this is a bit wider an area.

Hope this makes sense as I am writing it on the fly so to speak.

Regards

Daryl Mather
www.strategic-advantages.com

Risk-based work selection is a good concept but the detailed version is rather new in our case.

The simple format we use for plant work management is as follows:
1) Job priority 1 if there is a HSE and/or economic(production)loss and therefore need to eb attended immediately eg plant/equipment trips
2) Job priority 2 if there is a potential HSE/economic loss and need to be attended within 24 hours eg equipment problems found during prev/pred maint
3) Job priority 3 if no HSE/economic loss and need to be attended within 2 weeks (to allow for proper planning & scheduling) eg normal maint works
4) Job priority 4 for low-valueded/general works which can be attended within 1 months if cannot be eliminated
5) Job priority 5 for turnaround/shutdown works

From the above, backlog jobs are deduced, which can be good or bad in nature. Good backlog jobs are active work orders which have been properly planned and scheduled ie ready for execution by maint crafts. Active backlog jobs should be about 2-4 crewweeks to make jobs ready upfront for craftmen (front loading concept).

Bad backlog jobs are inactive work orders which should be expedited. Work orders become inactive due to various work constraints as follows:
1) operational constraints due to hot work or eqpt shutdown required to execute the works
2) non-availability of resources and/or services on the maintenance side eg materials, manpower, equipment, technical expertise

The above work order user status can be inputted into CMMS so that listing of work orders can be done easily eg during plant trips and no need manual recording.

Risk-based work selection should be applied rigorously to jobs with priority 1, 2 and 5. Turnaround works should be selected properly based on risk matrix if the aim is to optimize its interval and duration. Turnaround interval can be extended if the risk is acceptable and overhaul frequency for certain machineries can be extended as well. Certain works eg spared equipment can be carried out just before or after turnaround or evenly distributed during normal operations.

At the end, risk-based work selection should be able to prioritise works, add value to works done and prevent unnecessary works from being executed (maint prevention).

Hope to view your RBIM procedure. TQ