Reliability centered maintenance
Download as PPTX, PDF2 likes1,512 views
The document discusses methodology for maintenance, specifically preventive maintenance. It describes four main functions of maintenance as maintaining, keeping in existing condition, preserving, and protecting from failure or decline. Preventive maintenance is classified as either corrective or preventive. Preventive maintenance aims to prevent or mitigate failures from occurring and can be time-directed, condition-directed, or failure-finding. The document also discusses reliability centered maintenance and its features, including preserving system function over equipment. It outlines the seven-step methodology for implementing reliability centered maintenance on systems.
Reliability centered maintenance
- 1. Methodology
- 2. Functions of maintenance • To maintain • Keep in existing condition • Preserve, protect • Keep away from failure or decline
- 3. Classification of maintenance Preventive maintenance – Corrective maintenance Function of preventive maintenance Prevent/ Mitigate the failure occurrence Detect onset of failure Discover the hidden failure Do nothing, because of valid limitation
- 4. Preventive maintenance categories Time directed- The equipment is restored periodically at fixed time intervals. The intervals may be specified like clock time, cycles, and calendar days, seasons or prior to some events. Overhauls are preferred methods for this. Condition directed-find some parameter which co-relates to incipient failure. By measuring this parameter over the period of time onset of the failure is detected. Vibration, temperature etc Failure finding- Many equipments or systems may experience the failure and in normal course of operation no one would know that failure has happened. Backup system, emergency systems and infrequently used systems may feel this. These are called hidden failures. These systems are periodically operated just to check if it is in working order. Run to failure RTF- Doing nothing for preventive maintenance deliberately because of-1. We can’t find any PM task that will do any good irrespective of how much money we spend. 2. The potential PM is costly than repair when it fails and has no safety impact. 3. The equipment failure is too low on priority to provide PM as per budget.
- 5. CD (condition directed) Versus FF (failure finding) The equipment goes into the operation at TA and runs flawless till TB. At TB the failure mechanism initiated. If the failure continues to develop it can ultimately reach to TC where equipment may fail. At TD the equipment will be repaired and recovered to service. If TC is well beyond the TB we have time to conduct CD. If TC >> TB , CD possible. But if TB =TC (Electronic items) no time for CD. If TC=TD , The operator knows that something is wrong that is failure mode is evident. but if TD>TC the failure is hidden. To summarize between TB and TC we can define CD task, between TC and TD we can define FF task if failure mode is hidden
- 6. PM practices and Myths Most of time it was observed that its hard to find the origin of PM activities. Failure prevention- There is myth that all failures can be prevented and this motivates us to major/ minor overhauls without any understanding of failure mechanism. Which wastes a lot of money and becomes even counter predictive in lots of cases due to arising human errors? Experience- Most common answer given to justify the PM is heard experience, as it has been done from 15 years so it must be good. Experience is great but has to be used in a logical framework. Judgment- PM Activities initiated by judgments like, I think, this might be good, I think but I am not sure. If these judgments gone wrong they lead to big damage and money waste. There are some more like recommendation from OEM, regulations etc.
- 7. Reliability Centered Maintenance In 1960, 747 aircraft was manufactured in Boeing factory. The licensing authority recommended the extensive preventive maintenance which was 3 times than earlier 707 as 747 was 3 times bigger. It became economically unviable to operate by airlines. United airlines created a group of specialists who conducted a complete review of why maintenance was done and how it should be accomplished. This new technique was defined as MSG-1 (maintenance steering group). These ideas were adopted by Navy for their all aircrafts and they called it Reliability centered maintenance. In 1983, Electric power research institute (EPRI) studied these to apply for nuclear power plant.
- 8. RCM Features Preserve system function (and not equipment)- we must preserve the system function and try to know what is the expected output of it. Preserving the output is primary work. We must not assume an equal priority to every item of the function. Identify the failure modes that can defeat the function- functional failures come in many shape & sizes, it’s not always a yes or no. A pump supplying water at certain pressure to a boiler system. The system loss of fluid can be 1- minor leak, dripping, 2- A leak which will drop the pressure beyond a limit to create negative impact on the system, but system still operating. 3- Total loss of fluid supply. So a single function gives rise to 3 distinct failures. Priorities function needs via failures modes- since our aim is to preserve function, all functions are not created equal. Thus we need to prioritize the importance of failure modes for assigning the resources. This is done by passing three ties decision tree for each failure mode. Select applicable and effective PM task- address all failure modes in their prioritized order to identify the candidate PM actions. After listing all activities each has to be judged to be applicable and effective. Applicable means if task is performed it will prevent, detect or discover the failure. Effective means it is worth to spend the resources for conducting PM.
- 9. RCM Methodology- RCM requires a comprehensive system analysis to implement the four features. System analysis is a seven step process developed from experience as a convenient way to delineate the required information.
- 10. 1.System selection and information collection A plant or facility has so many systems. For selecting the system 80/20 rule can be applied. Some basic questions can be asked to select the system. Like.. A) cost of corrective maintenance actions. B) Number of corrective maintenance actions. C) Number of hours attributed to plant outages. Answers of the above questions will guide us to know who are the bad actors in the system and need the immediate attention. After selecting the system, information like drawing, design details, block diagrams, vendor manuals, and equipment history etc must be collected.
- 11. 2.System boundary definition There must be precise knowledge of what components has been included in the system so that an accurate list of components can be prepared and there is no overlapping of systems. Boundaries will be important in determining what comes in and what goes out of the system.
- 12. 3.System description and functional block diagram System description- key functional parameters, redundancy features, protection features and key control features are be recorded. At this level if the system is very big, can also be divided in subsystems. Functional block diagram- from the knowledge of description a simple block diagram needs to be developed. it will help in quick analysis. IN/OUT interfaces- interfaces are to be defined at the block diagram. System may have many in/outs so interfaces must be defined. System work breakdown structure (SWBS)- list of all components of the system and their interfaces, related drawing are to be recorded. Equipment history- equipment history available will be of great help so it must be recorded as well.
- 13. Block diagram of system
- 14. 4.System function and functional failures Sine the RCM emphasizes on preserving the system so now at this stage a complete list of system functions must be prepared. After preparing the list it can be listed out that how these systems can fail or what will be the failure mode.
- 15. 5.Failure mode and effect analysis Failure mode- this is a brief description of what is wrong like jammed, worn, frayed, cracked, split, open, torn and so forth. Failure effect- it is the consequence of failure. It can be analyzed on local, system and plant level. From the understanding of failure modes and their effect on different levels a decision has to be taken further. In one system different failure modes may have same failure effect. Failure mode must depict a problem that can be realistically addressed with PM. For example we never do a PM electronic chip. Redundancy- if a failure mode can be preventing from redundancy the priority will not be given to that failure mode.
- 16. Now since we have listed out all the functions of the system and failures, we will try relate the functional failure with the system components. We can use the adjacent matrix for this purpose. Here will identify the components which have potential to crate one or more failure modes. As per the matrix all components are related to one or more failures. If any equipment is not related to any failure, that means either there is mistake in analysis or we have discovered a component that plays no useful role in system.
- 17. 6.Logic (decision) tree analysis The failure modes which have passed the earlier screening will again be checked against the logic tree to allot the resources. When the LTA is concluded every failure mode is either classified as A,B,C,D/A,D/B, or D/C. The failure modes in A or B either hidden or evident have priority over C. Priority can be as below. 1. A or D/A 2. B or D/B 3. C or D/C The failure modes in bin C are usually left for RTF or for status only instruments if such is possible.
- 18. 7.Task selection After gathering all the information about the failure modes, effects and checking them on logic tree we arrive to failures modes for which PM is to be done. The PM task should be applicable and effective. Applicable means a task which can mitigate the failure and effective means cost effective among all other options If no applicable task exist, then only option is RTF If the cost of PM exceeds the cumulative cost of associated failure, then effective task option will be RTF A below roadmap can be followed to further decide the PM task
- 20. Task Interval & age exploration Selection of correct interval at which to perform the preventive maintenance task is the most difficult job. For deciding the task interval we need to have detailed knowledge of failure mechanism and age reliability relationship. Based on the available information we always take a conservative decision regarding the task interval. But after the we need to continually monitor and try to improve our knowledge about failure to explore the age-reliability relation. Every time we go for performing the TD task we need to inspect the condition, take photograph and document it for further study for evaluating the decision of task interval. Task interval can be increased or decreased based in the input.
- 21. Age exploration in United airlines Age exploration of hydraulic pump is united airlines. Initially overhaul started at 6000 hrs and in 4 year through age exploration the interval increased to 14000 hrs. At the same time the premature removal has decreased substantially.