RCM in Oil Storage RCM in Power Distribution RCM on Railways RCM on Oil Platform RCM in Power Generation RCM in General Industry RCM in Shipping RCM on Floating Hotel RCM in Paper Industry


Download eBrochures

1.       Abstract

2.       Introduction

3.       Determining Maintenance Task Intervals

1.      
Abstract


This paper sets out some of the most common myths, mindsets and mistakes that are made when establishing maintenance task intervals.

There are four basic types of maintenance that can be applied to equipment. The task intervals for the different types of maintenance are dependent on different factors – these factors are often poorly understood. This lack of understanding is a common cause of poor operational reliability and availability.

Successful reliability growth and performance improvement is all about ‘doing the right maintenance on the equipment’. Determining the ‘right maintenance’ takes time and resources.

Experience has shown that the only way to make real, lasting improvement in the maintenance arena is by:

The most common myths, mindsets and mistakes that  are made when establishing maintenance task intervals are summarised in the following paragraphs; a full explanation is given in the subsequent sections of this paper.

There are four basic types of maintenance that can be applied to equipment. The task intervals for the different types of maintenance are dependent on different factors – these factors are often poorly understood. This lack of understanding is a common cause of poor operational reliability and availability. [explanation]

A common statement that maintainers make is “We need to check our critical equipment more often than our non-critical equipment”. This sounds like good ‘common sense’ but is, in fact, wrong for On-condition maintenance. [explanation]

A common statement that maintainers make is “It doesn’t fail so often, therefore, I don’t need to check it so often”. This sounds like good ‘common sense’ but is wrong for On-condition maintenance. [explanation]

A common statement is “We monitor our equipment MTBFs carefully so that we can determine how often we should overhaul/replace equipment”. In fact the task interval (ie the fixed interval at which the scheduled restoration or discard task is carried out) is determined by the “life” of the equipment. Crucially, “life” and MTBF are different. [explanation]

Many maintainers think that there are just three types of maintenance: predictive, preventive and corrective – ie they mistakenly think that On-condition tasks (ie predictive maintenance) and Failure-finding tasks (ie detective maintenance) and are one and the same; they are completely different. [explanation]

The belief that collecting failure rate data leads to better maintenance is, in most instances, a myth. The data we need is rarely available and so key decisions about maintenance have to be made in the absence of hard data.  [explanation]

Organisations that rely heavily on protective systems frequently reduce the maintenance carried out on them in order to reduce overall spend; they assume that the protective systems will operate when required. These systems can and do fail; organisations may be vulnerable to serious consequences if the protected function subsequently fails. Maintenance spend must be directed to where it will do the most good. [explanation]

2.      
Introduction


The last 20-30 years have been characterised by massive technological change and most industries have responded by investing heavily in automation and technology to reduce headcount, improve product quality, reduce unit price and improve safety and environmental integrity etc. The net result is that organisations are increasingly reliant on their assets to perform when required.

In some organisations, equipment failure is becoming increasingly intolerable and the consequences of failure can seriously affect safety and the environment or be expensive in terms of lost production or customer service. Some failures are sufficient to threaten the financial stability of the organisation or even force it out of business.

In other organisations equipment failure is much less severe but can still adversely affect profitability or customer service.

Regardless of the industry sector, organisations are striving to increase cost effectiveness – in most organisations improving equipment reliability is the key to overall performance improvement and cost effectiveness. Successful reliability growth and performance improvement is all about ‘doing the right maintenance on the equipment’.

Many regard maintenance as ‘applied common sense[2]’ – in many respects this is true but only if the proponent of the ‘common sense’ actually understands what he or she is doing. This paper sets out some of the most common maintenance errors that many make when establishing maintenance task intervals.

The maintenance arena is littered with an assortment of myths, mindsets and mistakes which often mean that the resulting equipment maintenance does not achieve the desired outcome, is frequently flawed (sometimes fatally!) and is sometimes plain wrong.

Current management styles demand ‘instant results’ preferably via a ‘quick fix’. Frankly, it is a myth to think that quick-fixes work in the maintenance arena. Achieving reliability growth or performance improvement is neither quick nor easy; if it was, you would have done it by now!

Reliability-centred Maintenance  [RCM]  is an approach for determining the right maintenance for plant and equipment in its operating context.  RCM  is not a ‘quick fix’ but applied correctly, it can transform an organisation’s approach to maintenance and hence lead to substantial improvements in equipment reliability, overall performance and cost effectiveness.

There are four basic types of maintenance that can be applied to equipment:

RCM  Task Name


Description


Common Synonyms


On-condition


Check to see if equipment is failing


‘Predictive Maintenance’, ‘Condition-based Maintenance’ or ‘Condition Monitoring’


Scheduled Restoration & Scheduled Discard


Overhaul or replace equipment before it fails


‘Preventive Maintenance’ or ‘Scheduled Overhauls/Replacements’


Failure-finding


Check to see if equipment has failed


‘Functional Checks’ or ‘Detective Maintenance’


Corrective


Corrective action following discovery of failure or potential failure


‘No scheduled Maintenance’ or ‘Run to failure’.

Also includes work carried out following other types of maintenance


The task intervals for the different types of maintenance are dependent on different factors; these factors are often poorly understood by maintenance technicians and engineers. This is an area rife with maintenance myths, mindsets and mistakes.

Experience has shown that the only way to make real, lasting improvement is by changing the way in which people think (ie dispelling all the myths, mindsets and mistakes that have been ingrained in maintainers thinking over their working life) and providing them with an approach (such as  RCM ) that encompasses a structure whereby their changed thinking can be brought to fruition.

Part of the  RCM  process is to confirm that each maintenance task is technically feasible before it can be selected – the technical feasibility criteria ensure that task intervals are selected appropriately and establish the ‘common sense’ that is frequently lacking when maintenance tasks are determined in the absence of  RCM.

Applying  RCM  correctly takes both time and resources. However, for an organisation with high value assets, the investment required to get the maintenance right is often a drop in the ocean compared with cost of getting it wrong. Sadly, few maintainers are praised for getting maintenance right – many, however, are lambasted for getting it wrong.

RCM Specialists - Welcome

4.       Failure Rate Data

5.       Protective Systems

6.       Conclusion





[Contents]


[Contents]


3.
Determining Maintenance Task Intervals

Mistakes in determining maintenance task intervals are common (usually because the factors that determine the task intervals for the different types of maintenance are poorly understood).

A common statement that maintainers make is “We need to check our critical equipment more often than our non-critical equipment”. This sounds like good ‘common sense’ but is, in fact, wrong for On-condition maintenance. An on-condition maintenance task entails checking equipment to see if it is in the process of failing and only carrying out a corrective maintenance action if it is found to be failing.

For example, a bearing may be checked to see if it is noisy (ie indicating that it is failing) and if it is found to be noisy, arranging for a new bearing to be fitted (ie the corrective action). The task interval is determined by the time taken from the bearing becoming noisy to the point when it, say, seizes; this period of time is referred to as the P-F interval (or the failure development period). So if the P-F interval for the bearing is two weeks, then the bearing must be checked for noise at intervals less than two weeks (this is because the task might be carried out, say, the day before the noise develops and so the equipment would be left in service – the task must be repeated within the P-F interval if the failing bearing is to be predicted). Crucially, the task interval is not determined by the criticality of failure.

Similarly, a common statement that maintainers make is “It doesn’t fail so often, therefore, I don’t need to check it so often”. Again, this sounds like good ‘common sense’ but is, once again, wrong for On-condition maintenance.

Continuing the example of the noisy bearing, there is a temptation after, say, 18 months of weekly checks without any sign of imminent bearing failure to decide (incorrectly) that the task interval can be safely extended to, say, two weeks. After, say, a further 18 months of two weekly checks without any indication of imminent bearing failure the decision is taken to increase the task interval again to, say, one month. By extending the task interval the cost of the on-condition maintenance has been significantly reduced but the likelihood of the task predicting the failure is also significantly reduced. When (after, say, a further year) the bearing eventually fails and the consequences are suffered (because the task failed to predict the failure), the credibility of on-condition maintenance is questioned and even considered to be a waste of time. The task must be carried out within the P-F interval if the failing bearing is to be predicted). Crucially, the task interval is not determined by how often the bearing fails

Another statement that is common is “We monitor our equipment MTBFs carefully so that we can determine how often we should overhaul/replace equipment”. In fact the task interval (ie the fixed interval at which the scheduled restoration or discard task is carried out) is determined by the “life” of the equipment.

In essence, scheduled restoration and scheduled discard tasks prevent failures occurring by either restoration or replacement before an age-related failure occurs. Failure patterns A & B suggest that (if the initial period in A is ignored) the equipment is more likely to fail after it has reached a certain ‘age’ (ie the point at which the conditional probability of failure starts to rise rapidly). In  RCM  this ‘age’ is referred to as the “life” of the equipment (or sometimes the ‘useful life’). For failure pattern B, the frequency of failure typically peaks at some point after the end of the “life”; the point at which the frequency of failure peaks is sometimes referred to as the ‘average life’ or the MTBF.

Crucially, “life” and MTBF are different figures and if they are confused and the MTBF is inadvertently used to determine the task interval for a scheduled restoration or discard task, then many failures will occur before the task is carried out.

For example, on a petrol engine car it is normal to replace the spark plugs every, say, 12,000 miles as experience shows that starting, performance and emissions may be adversely affected if the plugs are not replaced. The actual MTBF (ie the ‘average life’) may be significantly longer than 12,000 miles.

Many maintainers think that there are just three types of maintenance: predictive, preventive and corrective. In doing so they have made the common mistake of confusing On-condition tasks (ie predictive maintenance) and Failure-finding tasks (ie detective maintenance) and thinking that they are one and the same; they are completely different.

On-condition tasks involve checking to see if the equipment is ‘failing’ so that an imminent failure can be predicted; as described above, the task interval is determined by the P-F interval. A Failure-finding task, however, is checking to see if the equipment has already ‘failed’; in this instance the task interval is calculated formulaically and does depend on the criticality of failure and the failure rates of the equipment concerned. The aim of a Failure-finding task is to improve the availability of the equipment (so that it is not in a failed state when required).

4.
Failure Rate Data


Many organisations have a (sometimes obsessive) desire to collect data so that they can calculate equipment failure rates in order to iterate to an even better maintenance schedule in the future. This myth is promoted by many in the maintenance arena (often because they have a vested in interest in encouraging organisations to collect historical data) but the usefulness of the information gathered is dubious:

With the exception of failure-finding tasks, the failure rate of equipment is not used to determine maintenance task intervals. The failure rate is, however, used to determine whether or not maintenance is worth doing and in determining spare parts stocking policy.

In summary, therefore, the belief that collecting failure rate data leads to better maintenance is, in most instances, a myth. In practice serious decisions about maintenance have to be made with inadequate hard data simply because the data that we need is rarely available.  RCM  provides a framework to highlight what information is needed and a structure to enable  RCM  Analysis groups to make the necessary decisions about maintenance task intervals (in particular) in the absence of hard data.

5.
Protective Systems


In modern, high risk process plant it is not uncommon for the maintenance of the protective systems to account for more than 30% of the maintenance labour spend on site; the reason for this is simply the number of these systems that have been installed (and need on-going maintenance).

Organisations that rely heavily on protective systems frequently make the mistake of assuming that their protective systems will act when required (ie if the protected function fails) and reduce the maintenance carried out on the protective systems to save money. However, protective systems can and do fail.

Over the years, the introduction of technology solutions and automation into many organisations was primarily focused on reducing product unit cost (largely by increasing throughput and reducing headcount - particularly production personnel). This transition has been successful in many industries and most organisations now rely very heavily on the installed technology and automation.

In addition to reducing product unit costs, the wide-scale application of technology and automation has done much to improve safety and environmental integrity. Many high-risk industries are very reliant on automation to ensure safety and environmental integrity - the risk of major incidents occurring has been reduced substantially by the introduction of protective systems (which are often configured in ‘layers’ so that failure of any one protective system is protected by another protective system etc).

The large population of protective systems installed and the level of protection provided does come with a risk - complacency on behalf of the maintainers (which is further compounded when organisation are under financial pressure to reduce spend). Automation in general and protective systems in particular can and do fail; some modern control systems are designed to identify and sound an alarm if there is a failure in the protection system but many do not. In practice, organisations may well be unaware that protective systems are in a failed state which, in turn, makes them very vulnerable to serious consequences if the protected function subsequently fails.

There have been some significant headline industrial accidents that have occurred because protective systems have been in a failed state when called upon to operate; several of these incidents have been as a direct result of complacency with regard to protective systems and some have been attributable to maintenance cost-cutting.

RCM  allows organisations to determine the maintenance requirements of the protective systems so that the combined probability of the system being in a failed state when the protected function subsequently fails is reduced to a tolerable level. This ensures that maintenance spend is directed to where it will do the most good.

6.   
Conclusion


The drive to improve cost effectiveness has led organisations to focus on increasing equipment reliability so as to improve overall performance (and hence cost effectiveness). Successful reliability growth and performance improvement is all about 'doing the right maintenance on the equipment'. There is a great temptation to improve cost effectiveness by reducing maintenance budgets - this works in the very short term but not in the medium and longer terms.

The maintenance arena is littered with an assortment of myths, mindsets and mistakes which often mean that the resulting equipment maintenance does not achieve the desired outcome, is frequently flawed and is sometimes plain wrong. In particular, it is a myth to think that quick-fixes work in the maintenance arena. Achieving reliability growth or performance improvement is neither quick nor easy; if it was, you would have done it by now!

Reliability-centred Maintenance  [RCM]  is a proven approach for determining the right maintenance for plant and equipment in its operating context. With its beginnings in the demanding civil aviation industry,  RCM  is not a 'quick fix' but applied correctly, it can transform an organisation's approach to maintenance and hence lead to substantial improvements in equipment reliability, overall performance and cost effectiveness.  RCM  optimises the maintenance for the on-site equipment and in so doing ensures that money spent on maintenance is spent where it will do the most good.

Applying  RCM  correctly takes both time and resources. However, for an organisation with high value assets, the investment required to get the maintenance right is often a drop in the ocean compared with cost of getting it wrong. In brief, the return on investment for  RCM  is substantial provided that the assorted maintenance myths, mindsets and mistakes are dispelled and the right maintenance is implemented.

This paper was written by Simon Deakin and Steve Bailey of Mutual Consultants Ltd.

See also Maintenance Myths, Mindsets and Mistakes Parts 2 & 3, Improving Operational Reliability & Availability and The Maintenance Arena respectively.

[1]  RCM  – Reliability-centred Maintenance. Short-form definition: “A process used to determine the maintenance requirements of plant and equipment in its operating context”

[2] A dictionary definition of ‘common sense’ is: sound practical judgment that is independent of specialised knowledge, training, or the like; normal native intelligence.

[3] HL Resnikoff, Mathematical Aspects of Reliability-centered Maintenance, 1978

[Contents]


[Contents]


[Contents]


[Contents]


[Contents]


Go to eBrochure Go to eBrochure RCM Specilalists RCM Specilalists RCM Specilalists RCM Specilalists RCM Specilalists RCM Specilalists RCM Specilalists