RELIABILITY-CENTRED MAINTENANCE - An Introduction - CONTENTS PAGE 1. THE CHANGING WORLD OF MAINTENANCE 1 2 MAINTENANCE AND RCM 3 RCM: SEVEN BASIC QUESTIONS 3.1. Functions and Performance Standards 3.2 Functional Failures 33. Failure Modes 3.4 Failure Effects 35 Failure Consequences 36 Proactive Tasks 3.7. Default Tasks 3.8 Tho RCM Task Selection Process 4 APPLYING THE RCM PROCESS. WHAT RCM ACHIEVES ay Alacon1 The Changing World of Maintenance ‘1 Over the past twenty years, maintenance has cham sei per haps more so than any other management discipline. The increase ia the number and squipment onl bsldings} to be maintained, more complex designs, new changes are due to 4 bi variety of physical assets (plant that ned ‘maintenance techniques and changing views on mainte Maintenane These includes pidly grow ination and responsibilities, isalso responding tachanging: sspectations, wareness ofthe extent to which equipment failure affects safety and the environ ‘ment, a growingawarenessof theconnestionbaetween main- tenance and product quality, and increasing pressure to achieve high plant availability andl to conta costs attitudes and skills inall bran ‘ches of industry tothe lini. Maintenance people are havi These changesaretestin, to adopt completely new ways of thinking and seting, as s. Abthe same time the limita ns are hacoming increasingly ineers and as man tions of maintenance sy apparent, no matter how much they are computerised In the face ofthis avalanche of change. me where are seeking « new approach to maintenance. The wan 1 avoid the False starts and dead encls that always accompany majorupheavals. Jasteudshev seekastrate ic framework that snshesises the ne’ coherent paitern, so that they can evaduate thea sensibl aan appl shove likely 10 be of moss value tothe and thett companies, This paper describes « philosophy that provides such Framework. Its or RCM Titisapplied correctly, RCM transforms therelationships henveen the undertakings which use it. their existing physi sl assets and the people who operate and maintain thse alled Reliability centred Maintenance assets, It also enables new assets to be put into effective service wit lowing pang starting with s look at how maintenance by the past fifty years Since th traced through three generations, RCM is rapidly becom eat speed, confidence and precision, The sriphs proviclea bref introduction to RCM, 1930's the evolution of maintenance can he cornerstone ofthe Third Generation, butthis genera First and Second Generations only be viewed in perspective in the The First Generation The First Generation covers the period upto World Wa I In those days industry was not very highly mechanised, so downtime did not matter much. This meant that the pre vention of equipment failure was how high priority inthe minds of mast managers. At the same time, most equip. ‘ment was simple and generally over designed. This made st reliable an eass to repr, As ares, there wais no need For systematic maintenance of any somt icing and heyond simplecleaning 8 lubrication routines, The need for skills was also lower than itis today The Second Generation Things changed dramatically during World War Il, War tinte pressures inereased the demand for goodsofall kinds he supply of industrial manpower cropped sharply This led to increased mechanisation, By the 1950s mi chines of all types were more aumerous and more con to depend on them As this dependence grew. downtime came into sharper ‘that equipment fail sand shoul be prevented, which led in turn to the concept nev. In the 1960's, this consisted xquipyment overhauls Gone al fived intervals, whi ple. Inulastry was fircus. This lel to the i could rainy 0 The eostof maintenance also started to rise sharply rel tive to other operating costs. This led tothe gnowth of mia enanice planning and control sxstems. These have helped greatly tobring maintenance undercontrol.and are now an part of the practice of maintenance. Finally the amount of espital tied up in fixed sssets together with aashaup increase inthe cost ofthat capital led people tostatt secking ways im which they covald maximise the life ofthe assets. The Third Generation Since the mid seventies, the process of chang inindustry has gathered even greater momentum, The changes ca be classified under the hi lof mew expectations, new research and new sectnigues + New expectations. Figuee | shows how expectations of rnaintenance have evolved, Downtime has always alfee teal the productivecapability of physiealassetsby reduc ing output, increasing operating costs ancl interfering Ty the 1960s and 1970's, this manufactur ing and transport sectors, The effeets of downtinte have been aggravated by the worldwide move towards just in time inventory management - stock levels in general have heen reduced to the point that minor equipment sorts of logistic support systems. In recent times. the growth of with customer servic ‘was already a major concern inthe minin failures can now have & major impact on » fautonnation has meant that redial anal evallabiiry havealsobecomekey'issucs insoctorsas diverse ashealth ‘care, data processing. telecommunicationsand building anagement, Figure 1 Growing expectations of maintenance © sas9 JALADON© 10 ALADON. + New seseeiroh: Quite apart tron + New seviniguues: The Greiter automation alse means that more aad more failures affect our ability to sustain satisfactory guuatrs standards. This applies as much to standards of service to product quality. For instance, eaipmient failures affect climate control in)suildings and the pune tality of wansport nenworks as much as they interfere as itd with the consistent achievement of specified tolerances, More and more failures have sertous sagen or enn ronmental consequences, ata tite when standards in thes. the point isapproachit Form to soviety's safety and environmental exp orthey cease to operate. This adds an orderof a wo se on the integrity of our physi assets one that goes heyond cost and becomes. sintple matter lepender of organisational survival, Aithesamtetimeas ourdependence on physicalassets is growing, so 109 is their casr— 10 0p erate aoa 10 OW To secure the maximum returnon the iavestarest which they represent, they must be kept working efficiently for lng as we want them to, Finally, the cose ofmaintenanceitself is still isin sa proportion of total expenditure In some industries, itis now the second highestor even the highest element of operating costs. As a estlt, im only thirty years it has moved Sram almost nowhere 40 the top of the league as a cost control priority eater expectations, new research is changing many ofourmest basic elie® about age and failure. In particuler, itis appsrent thet there is less and less connection between the operating (oF most assets and bow likely they are to fil ure 2showshowth sarlist view of failure was sim ply thatas things got okler they were more likely to fal A growing awareness of “infant mortality” Io to wide sprecd Second Generation belief inthe “satu” curve However, Thitd Generation research hus revealed that notone or two but sé failure patterns actually occur in practice. This is discussed in more detail later but it profound is been explosive new maintenance concepts and techniques, Hundreds have been developed over the past twenty years, and nore are enierging every week. The chissical emphasis fon overhauls snd administrative systems has grown 10 include many new developments in a numberof differ cent fields. The new developinents include decision support toot, suc as haraed studies, failure modes and effects analyses and expert systems + new matnuenance techniques, such as condition monitor jrning equipment with a J major shifein organisational thinking ticipation, teamworking and flexibility A ragjor challe ays is Hot only 6 Fearn what these techniques ane, ut to decide which are worthwhile and which are not in their own organisations, Fwe make the ight choices, it owas par acing maintenance people sows ispossibletoimprove asset performance and atthe sume Zine contain andeven reduce the cost of matntenanee. swe make the wrong choices, new problems are ereated while existing problems only get worse The challenges facing maintenance Ina outshell, the key challe modern mainte facing nance managers can be summarised as Follows: + to select the most appropriate techniques + to ceal with each type of failure proc + inorder to fulfil all he expectations of the owners of th assets, the users af the assets and of society as 8 whole’ + in the most cost effective and enduring fashion + with the active support and cooperation of all The peopl involved, The first industry to confront these ehallengy systema cally was internstional civilaviation. Inresponse to many ofthe new developments that are part of the Third Gene ration, this industry developed a ‘work for developing maintenance strategies, This frame ‘work is known withia aviation as MSG. and outside itas Reliability centred Maintenantee, or RCM. Since the carly 1980's, the Aladon network has helped users to apply RCM on more than 1000 industrial loca world — work that leo the develop #f RCM? in 1990, The rest of this paper introduces RCM in more det Part 2 explore ‘id goes om to devine RCM. Part 3 suntmarises the seven) RCM. (The process sum nurised in Part 3 this paper complies fully with 'SAF Standard IAIOM: “Evaluation Criteria for Refiabrtion- Centered Maintenance (RCM) Processes. Part 4 leseribes how RCM should be applied, and Part S outlines what it achieves. the meaning of the word “maintenance key steps involved in apply (on equipment failure ‘eater empphasison reliatil ity and maintainabbity LgFrom the engineering viewpoint, there are nwoelementst0 the management of any physical asset must he rain tained and occasionally it may also need to be modified The major dictionar continue (Oxford) or keep in an existing state Webster) ‘This suggests that maintenance means preserving some thing, On the other hand, they agree that 4 veo some thing means to changeit in some wey. The importance of inthe RCM decision process. this point et out t0 maintain something, what is it that letine muinrain as cause 20 thisdistinetion ise However When wwe wish to carve ta continue? Whit is th that we wish to preserve? The answer to fincus on maintenane {questions can he found in the fact that every physieal asset is put into servive becausesome fone wanls itdo something. In other words. they expect it Before RCM to ean be app tovany asset or syste, iis estab lish the System boundaries, cleathy define ils operat, context, and prepare a detailed pkmofaction. These issues are discussed at gr partof this paper biefly REM entails ask system under re + what are the functions and associated performauce standards of the asset in its present operating context? + in what ways does it fail to fulfl its functions? + what causes each functional failure? + what happens when each failure occurs? «+ in what way does each failure matter? + what can be done to predict or prevent euch failure? + what ifa suitable proactive task cannot be found? what systemt isto be analyse or length in Part 4of this paper. This scribes the RCM process itself Leven questions about the asset or Ww, 88 fllows’ These questions reviewed in the following paragraphs, 3.1 Functions and Performance Standards Bevore itis possible to determine what must he done 60 ensure that any physical asset continues to do what its users, wat ito do in its pres jelermine what its ws nt operating contest, we m ss want it do ut do what its users Want to start with, + ensue that ite This is why the first step in the RCM process entails d fining the fanetions of each asset ia its operating context \ users expect assets th he able todo ca ther with the desired standardsof performance. What split into v0 categories which summarise why the asset was first place, This category covers issues pro such as speed, output, carrying er stora uct quality and customer serve \liaintenance ACM: Seven Basic Questions and RCM 2 to lulfila specific function or functions. So it follows that when we maintain an asset, the stat wish to preserve pnuist be one in which it continues to do whatever its users want itt do Maintenance: Ensuring that physical assets continue to do what their users want them to da Whar the users want depends on where and how the asset is being used (the operating context), This leads tothe fol lowing definition of Reliability centred Maintenance: Reliability-centred Maintenance: a process used to determine whit must be done to ensure that any ‘physical asset continues to do whatever its users want it ta do in its present operating context’. + second fictions, which recognise thal every asselis expected to do:more than siraply full its primary fun tions. Users also have ex ations in areas such as safety, environmental compliance, control containment, ity, economy, protection, effiet ency of operation, and even the appearance ofthe asset ‘comfort, structural inte ‘The users of the assets are usually in the hest position by farto know exaetly what contribution the physical and finanefal well being of the org tava whole, so its essential that they RCM proce seh assel makes to anisation re involved in the Frown the sutict 3.2 Functional Failures ‘The objectives of maintenance are defined by the fune tions and associated performance expectations of the asset But how does maintenance achieve the The only occurrence that is likely to stop any asset per Forming tothe stanlard required hy its users is some kind of failure. This sug: ichiewes its objectives hy adopting suitable approach to the nize ment of failure, However, bef blend of failure mas objectives? sts that maintenance wwe can apply a suitable wement tools, weneed to ideatify what iilures can aceur, RCM does this at 0 levels + firstly. by identifying what circumstances amount toa failed state + then by asking what events can ease thensset to get into a failed state Inthe worldof RCM, failed states are knowns fametional Sailures secause they occur when an asset is tmable 10 full et faction to a standard of performance which is acceptable to she nscr. In adaltion to the total inability 10 this defin where the asset still funetions but atan unaveeptable level ‘unetivn, jon encompasses partial failures ‘ot performance (inchudin rol sustain eptable levels of quality or aecura © sas9 JALADON© 10 ALADON. 3.3 Failure Modes AAs mentioned in the previous paragraph, once each fine tional failure has been identified, the next step is to try to identify all the each failed « : ‘occurred onthe ss we, These event led failure modes. asonably likely" failure modes include those thet have stinginthe some context, failures that are currently being prevented ne or similar equipment oper by existing maintenance regintes, and failures that fave not happened yet but are considered ta he real possibilities in the context in question. Most traditional lists of failure modes incorporate fail However the list should include failures caused by human caused by deterioration or normal wear and tear. errors (on the partof operators nd maintainers) and design flaws so that all reasonably likely causes of equipment failure ean be identified andl dealt with appropriately. Iris also important to identify the cause of each failure in enough detail for ittabe possibleto identify an appropriate failure management policy 3.4 Failure Effects ‘The fourth step in the RCM provess entails listing failure effects. which describe what happens when each failure nnioxle ageurs, These descriptions shovld include all the information neede to support the evaluation of the conse quences of the Failure, such 3s! + what evidenee (if any} that the failure has occurred + in what ways (iF any) it poses a threat to safety or the environment hat ways (if any) itaffeets production or operations + what physical damage (if any) is caused by the failure + what mist be dome to repair the failure 35 Failure Consequences A detailed analysis ofan average industrial undertaking is likely to y lure modes. Each oF these failures aff tion in some way, out in each ease, the effects They may affect operations, They may also affect product ‘quality customer service, safety or the environment They will all ake tine The: to which we try top a failure has serious consequences, we ane Tikely to #0 t0 hs to try to avoid it. On the other hand, if it has Title oF no effect, then we may decide to dono routine «!Inbrication, sognises that the ‘and ten thowsand possible find cost money to psi influe the extent nther words, if ices most stro 1 each fail utlen maintenance heyond A great strength of RCM is that it consequences oF failures are far more important than their technical characteristics, In fact, it recognises that the only reason for doing any kind of proactive maintenance is not toavoid stto reduce the consequenceso? failure, The RCM process classitiesthese ilures perse, but oavoidorath consequences into four gromps, 3s follows: + Hidden failure consequences: Hikien failures have 80 Girect inapact, but expose the organisation to multiple failures with serious, often catastrophic, consequences. * Safety and environmental consequences: A {ailure has safety consequences if it could injure oF kill some fone. Ithas a corporate, rental sandr + Operational consequences: © ‘silure hes operational consequences if it affect quality, customer vironmental consequences ifiteoud breach vl, national or international environ production (output, product sosts imaddition to the direct cost of repair + Non-operational consequences: Fvisien\ ‘aihures that fall into this category affect neither safety nor opens tions, so they invalve only the direct cost of repair ‘The RCM process uses these categories as the basis of a stat By forcing failure mode in terms af the above categories, it integrates. framework for maintenance decision makin structured review ofthe consequences ofeach the operational, environmental and safety objectives of Thishelpstobring safety anultheenviranment into the mainstream of maintenance management The consequence evaluation process also shifts empha sisaway from the idea that af failures are had snd must be prevented. In so doing, i ocuses attention on the mainte set on the perform ofthe organisation, and diverts energy away from those thet have litle effect. It also encourages us to think m0 Droally about different ways of managing failure, rather thant concentrate only on failure prevention, Psiture nian mnent techniques are divided into two categorie nance activities that have most e + prowctive tasks: these are tasks undertaken befre a failure occurs, in orl into a failed state, They embrace what is traditionally knowa as “predictive” and "preve RCM uses the terms sohedlaled restoration, scheduled discard and on-condition maintenance fo prevent the tent front gett * default acrions: these deal with the failed state, and are chosen when itis not possible fo identify am effective proactive task. Default actions include farlure. nding, redesi 36 Proactive Tasks Many people still believe that the best way to optimise plant availability is to do some kind of proactive mainte ‘on a routine bass, Second Generation wisdom sug that this should comsist of overhaulls or component sements at fixed intervals. Figure 3 illustrstes the fixed interval view of failure Figure 3: Tre traditional view of failureBis hase ate reliably fora period *X°, and then wear out. Classical think tenable us to determine this fife and so make plans 0 take preventive action shortly before the iter is due to fail in onthe assumption that ntost items op. sts thal extensive records bout failure will fun This model is true fo and for som Inpartivular, wear-out charscteristies are often found where equipment comes into direct contact withthe product, Age certaintypes o complex items with don related failures are also often associated with fat corrosion, abrasion and evaporation, huipment in than it was thiny years ago. This has led to startling changes in the patterns oF failure. as shown in Figure 4 Figure 4:Six patterns ot failure The graphs show consltional probability of Failure against oper cal items ge For wide variety of electrical and mechani- + pattern A is the well known bathtub curve, It heins ‘of Failure (knosen as Jin mor fiy) followed by a constant or gracully inc ccondltional probability of failure, then a wear out zone + pattern B shows constant or slowly inereasing condi tional probability of failure, ending in a wear out zone lie sane as Figure 3), + pattern C shows slowly increas tlitional probabil ity of failure, but there is no identifiable wear out age, + pattern D shows low conditional probability of failure when the tem: is new or just out oF the shop, then 3 rapid increase to a constant level + pattern E shows a constant conditional probability of frilure at all ages (random fatlure) + pattern F tarts with bigh into mortality npping Wo a ing conditional prosability of constant or slowly’ ‘alow Studies done on civil aircraft showed that 4% ofthe items conformed to pattern A,2% t0 B.5% tC. 7% to D, 14% to Band no fewer than 68 so pattern F: (Phe aumber af ily the same as an industry. But there is no doubt that as assets times these patterns occur in aircraft is not necesss hhocome more complex, we see mone and more of patterns Band F.) ‘These findings contradict the belief that there is always, 4 comection between reliability and operating age, This belief led to the idea that the more often an itera is over hauled, the less likely it isto fai, Nowadays this is seldom true, Unless there is age limits do litle oF nothing to improve the reliability of Gominant age-related failure mode, comple tenis. In fet schealuled overhauls often mucreae foverall failure rates by introducing infant mortality inte otherwise stable systems, An awareness of these facts has led some organisations to abandon the idea of proactive maintenanee altogether. hth In fact, this can be the # foe For failures with minor consequences, Bul when the failure consequences are signific P 1, somerhing must be done to prevent or ict the failures, oral leest to reduce the comsequen This bringsus back tothe question ofprosctive tasks, As mentioned ealier, RCM div ides prowetive tasks into three categories, 2s follows + scheduled restoration tasks + schediled discard tasks + schediled on-condition tasks Scheduled restoration and scheduled discard sashs Schediled restoration entails rem facturing a compo- nent or overhauling an assembly at or before a specified ime, Similarly an item at orhefore a innit, regardless of its condition at ths scheduled discard entails discarding specified life lint, regardless of its condition at the time, Collectively, these two typesof tasks are nerally known as preveniive maintenance. They used to be by far the most willy used fort of proactive maintenance, How cover, for the reasons discussed above, they are much less widely used than they were twenty years go, On-condivion tasks ‘The continuing need to prevent certain types of failure andthe growing inability af classical techmigues to do so, are behind the growth of new types of f ‘ment. The majority of these techniques rely onthe fact that most fibures give some warning of the fact that they are about to occur. These warnings are known as potential Sallures, 2x6, se deed as identifaahle physicat condi tions which indicate that a functional failure iv abort to ecur or is tn the process of occurring The new techniques are used to detect potential failures so that ction can be taken t reduce or eliminate the com sequences that could oceur if they were to devenerate into funetional failures, They arecalled ov-comuition task, amd include al forms of evnaition-hasedmainsenance, predic © sas9 JALADON© 10 ALADON. Used appropriately, on condition tasks are a very good ‘way of managing failures, but they can also be an expen sive waste of time, RCM he made with particular confide tables lewis i this area to 3.7 Default Actions RCM recognises three major eategories of default action: + puilure- finding. Woibure-tingling entails checking biden ‘unetions to find out whether they have filed (as op- posed to the on coaulition tasks described above, which ifsomethingis faire The rapid yrowth in the use of built-in protective devices means that this ory of tasks is likely’ to becom sail checki a mainte nance mtnagement issue in the next ten years as cond tion monitoring has been in the last decade. RCM pro vides powerf il, risk-focused how often and by whom these tasks should be done vredenign-vedesign entails aking any one time change: tothe built-in capability ofa system. This includes mod iffcations to hardware and changes to procedures. (Note that the RCM process considers the maintenance require iments of each asset before asking whether itis necessary to change the design. This is because the main tenance person who ison duty Zed bus to mina the 2s in exists gaxiay, not what should be there or what might be there at some stay inthe funwe, However. ifit tans pires that an asset simply cannot deliver the desined por formance, RCM helps to focus redesiga efforts on the reall problems) sts, this 10 elfort to anticipate or prevent ilure modes ta which itisapplied, and sothose failures, are simply allowed to aceur and then repaired. This efault is also called rrr ro-favture, etal entails makin 3.8 The RCM Task Selection Process A great strength of RCM isthe way it provides precise and easily understood criteria for deciding whieh (if any) of the proactive tasks is seesicaaly feasible in any context and by whontthey should and if so for deciding how oft be done, Whether or not aprasetive task is technically feasible is governed by the secintcal characteristics of the task and ‘ofthe failure that iris meantto prevent. Whether itis 27h doing is governed by how well it deals with the conse quences ot be failure. Ia proactive task cant be found that is both fechnieally leasiblesnd worth doing, then suit he default ction must he takea. The essence of the tsk selection process is as follows 4 Applying the RCM Process Correctly applied, RCM leads to remarkable improvements in maintenance effectiveness, and often does so surpris ingly quickly. However, as with any fondamwental change nent project, RCM is mutch more likely to succeed task is worth doing + for hidden faites, a prouctis reales the risk of the ship flue associated! with that funtion toa tolerably low level. It sueh a task can sot fom then a ches fare finding tasks bye proseribed. [Fa suitable failure Finding task casnot be ision 4 thatthe ending on the find, then the secondary default fnem may have w be redesigned (e consequences of the multiple fail + for failures with sufi or emironmentad consequences, progetive hisk is only worth doing if it red ‘oF that failure on its own toa very low level indeed. it dyes nol eliminate altogether Ia task eanmot that recces the risk 10 a tolerable level, the dem must be redesigned or the process must be changed. + ifthe Failure has peat conse task isonly worth doing ifthe total cost of doing it over a period of tame is less tha the cost of the operational consequences and the cost of repair over the same peri ‘4, In other words, the task must be justifed o7 ome grands. Wi isnot justified, the initial default ocision is no scheduled maintenance. (i this. ices re still unacceptable * the risk found 1s, aprnaetive sand the aperstionsal conseqa then the secondary default decision is again redesign} + ifs failure has no-aperavionat consequences a prose ifthe cost of the ask overs is Tess than the cost of repair over the tive tsk is only worth doi period of tim same period. So these tasks must also be sieited on ceonomie- grounds It itis not justified, the initial default decision is ayuin 0 scheduled maintenance: iC ne the secondary default decision repair costs ute woh is once again redesign, ‘This approuch means that prostive tasks are only speci fied lor failure moxies that really need then, which in turn ' to suhstantial reductions in routine workloads, Less routine work. also means thet the remaining tasks are more likely tobe done properly. This together with the elisnina tion of counterproductive tasks leads to more effective intenanee. ‘Compare this with the tational approach tothe devel snanee policies, Traditionally, the mainte nance reqiirementso each asset areassessedin terms ats characteristics, without consid of failure, The resulting schedules opment oF main real or assumed techn he consequs are used forall similar assets, ag in without considering tat different consequences apply in different operating ‘contests, This rests il fof schedules that not because they are “wrong” in the technicsl sense, but hecause they achieve nothing. are wasted if proper attention is paid to thorough planning, how and by who the analysis is performed, auditing aa! imple entation, These issues are discussed in the following paragraphs,Priavitsing assets and establishing objectives art 5 ofthis paper expltins that RCM can improve organ isutional performance ina host of different ways, t and intangible, Tangible benefits include greater safety improved environmental integrity, improved equipment availability and reliability. htter product quality and cu tomer service andl reduced operating and maintenance costs, Intangible benefits include better understanding about bow the equipment works on the part of operators and ria tuiners, improved feamworking and higher morale RCM should be applied first t systems where it likely to yield the hi required in any oral of the above areas. [F these systems arenot self evident, itmay be necessary fo prioritise RCM. projects ona more formal basis. When this bas itis then essential 10 phan each project in derail Planning The successful lest returns relative to the effort pplication of RCM haps foremost on meticulous planning and pre reads First and per ation. The process are as follows: key elements of the planny + Define the seope and boundaries of each project + Define ancl wherever possible quantify the objectivesor ich project (now state and desired end state) + Estimate the amount of time (umber of meetings} reedded to review the equipment in each are + Identify project manager and facilitatorts) + Identify participants (hy title and by mane) + Plan uriaing for participants and facilitators + Plan date, time and location of each meeting + Plan management audits of RCM recommendations Review groups Weave seen how the REM process embodies seven basic ‘questions. In practice, maintenance people simply canal ‘answer all these questions on their owa, This is because many (iP not most af the answers can only’ be supplied by production oroperations people. This appliesespecially ta ‘questions concerning fail ture effects and fal junctions, desived pesTorman Forthisreason, areviewaf the maintenance requirensents of any asset should he done by snl teams that meluge a feast one person front the maintenanee funtion ancl one from the operations function. The seniority of the group fuet that they should have a thorongh knowledge of the asset under review rnrembets is Fess important than 1 ach RCM, The make-up of a typical RCM review group is shown in Figure 5. The use oF these oupy member should also have been trained in ment fous not anly enables ma to gain access to the knowledge and expertise of each rember ofthe group om a systentatic basis, but the mem hers learn great deal about how the asset works, Facilitators RCM review groups work under the gui ‘aclitators are the most important people in the RCM review process. Their tle is to ensure that: tance of highly specialists in RCM, Known ay facifitatars, The 7 Figure 5: & typical ROM rev 1 group + the RCM analysis is carried out atthe right level, that ol that no important items are overlooked snd that the results of the analysis are properly recorded system boundriesare clearly de + REM is correctly understond and applied by the group soup reaches consensus ina brisk and orderly fash- ion, while retaining their enthusiasm and commitment + the analysis progresses as planned sad Finishes on time Facilitators also work with RCM. project man Sponsors to ensure that each analysis fs properly planned sal and logistic support and receives appropriate man The onrcomes of an RCM analysis IF itis applied in the manner s ‘analy sis results in three tangible outee jested above, an RCM. 28. as follows: + schedlles to be done by the maintenance department + revised operating procedures for the operators of the asset list of areas where one-time changes must be male te the design of the asset or the way in which itis op. 10 deal with situations where the asset cannot deli the desired performance in its current configuration, Aless tangible ut very valuable outcome is tht participants inthe process start functioning much better as multi dis ciplinary teams after their analyses have been completed Auditing ‘fer the review has been completed for each asset, senior managers with overall responsibilty forthe equipment ust satisfy themselves that the review is sensible and defen sifsle, This entails deciding whether they agree with the definition of functions and performance standards, the identification of failure modes and the description of fail ure effects, the assessment of failure consequences and the selection of tasks Implementation Once the RCM review hasbeen auditedand approved, the final step is to implement the task. pro. time changes. The tasks and proved mented in & way that ensures that they will be clearly understood an! performed safely bythe people to whom they are allocited tures and one roust he docu: sas9 JALADON© 10 ALADON. The maintenance tasksare then fed into suitable high- and low frequency maintenance planni while revised opersti 5 Desirable as they are, the outcomes listed above should oly beseenasa meansto amend Specifically, they should enable the maintenance function to fulfil all the expecta tions listed in F ‘oP this paper: How they do so is summarised in the following paragraphs. nd control systems, rocedures usually incorpo ure Tal the begin + Greater safetys\ environmental integrity: REM sidersth failure mexte before eonsideringitseffecton operations ‘This means that steps are taken to snimse all identi able equipment related safety and environmental haz ards, ifnot eliminate theraltogether, By integrating sate ty into the mainstream of maintenance d RCM also improves attitudes to safety ety and environmental implications of every ision making. Improved operating performance (output, product quality and customer service): RCM ecognises thst al types of maintenance havesorie val idlesrles ‘anal for deciding which is most suitable in every situation. By doin: Forms of maintenance arechosen foreach asset, and that 0, ithelps ensure that only the most effective stitable action is taken in eases where maintenance ccsnnot help. This much more tightly focused! mainte nance effort leads to quantum jumps in the performance of existiag assees where these are sought. REM was developed to help airlines diay up main tenance programs for new types of aircraft Aejre they cater service. Asa result, itis an ideal way to develop such pr pecially comple. equipment ail- able. This saves much of the trial and ersor that is so Ollen part of the development of new main _grams—trial that is time-consuming and Srustrating. anc jetror that can be very costly ms for nen assece ‘or which no historical information is av ince pro Greater maintenance costeffectiveness: RCM con tinmally focuses attention on the maintenanc that bave most effect on the performance of the plant. This helps to ensure that everything spent oa mainte nance is spent where it will do the most goo! In addition, if RCM is correctly applied t0 exist ystems, i reduces the amount of routine work (in other words, maintenance tasks to he under- takenona cvericbasis) issued in 40% to 70%, On the other hand, if RCM is used to deve- lop a new maintenance program, the resulting scheduled schperiod, usually by \workload is much lower than ifthe program is developed! by traditional methods Longer useful life of expensive tems, ‘we ws carefully ‘Focused emphasison the useofon concitioa maintenance. rated into standard operating procedure manuals, Proposals fin dealt with by 0 orpro: pviications ject management funetion in most orgenisations. Vhat RCM Achieves 3 + A comprehensive database: Ass RCN review ends with a comprehensive and fully documented record of the nraimtenance requirements of all the significant assets used by the organisation, This makesit possible eaadapt to charging circumstances (sch as changing shift pet tems or new technology) without having t0 niaintenamce policies fom scratch, ILlso ensibles equips nent users 10 dernonstrate that their maintenance pro canis are bull ow rational fownations (the cir tel laters}, Finally, the in sheets reduces she of required by more and more formation stored on RCM wor feetsof sialf temover with ts attend loss 0 exper tence and expertise. An RCM review of the maintenance requitements of each assetalsoprovidesa much clearer view of the sis required fo maintain each asses and for deciding whut spares shun be held in stock + Greater motivation of individuals, especially people who are involved in the review process, This is necom panied by much wider “ownership” of maintenance pro’ lemsané their solutions. Italso means that solutiens are more likely to endure + Beter teamwork: RCM provises 2 common, easily un derstood technical any hing nce. This gives muaintesnance and operations peaple a better understanding of what mainte nance ean (and cannon) achieve aaxd what must be done to for everyone who has aay » do-with main achieve it All of these issues are pat of the mainstream of mainte agement. and many are already the target of i programs. A mayor feature of RCM is that it Dy-step framework fortackling aif of ther at once, and for involving everyone whe has. o do with the equipment in the process, RCM yields results very quickly. In act, ifthey are cor rectly focused and correctly applied, RCM analyses ean poy for themselves in a matter of months and sometimes even 4 matter of weeks. The process transforms both the perceived mai * the physieal as sets used by the organisation and the way in which the: maintenance funetion as a whole is perceived. The result prowides anclfective ste sanythi uirements 0 is more cost-effective, more harmonious and much more successful maintenance 1 Internationa Society of Automotive Enginsers: JA/OL value trom Enters for Noirs Contre Stantrance 1RCH) Fra fesse: Warten, Poesia, USA: SAE Pulications