How to de­velop an pre­dic­tive main­te­nance pro­gramme

DEMM Engineering & Manufacturing - - PNEUMATICS HYDRAULICS -

There are no easy so­lu­tions to re­duc­ing the costs of main­te­nance. The amount of time and ef­fort re­quired to se­lect pre­dic­tive meth­ods, that will pro­vide the most cost-ef­fec­tive means to eval­u­ate the op­er­at­ing con­di­tion of crit­i­cal plan sys­tems; es­tab­lish a pro­gramme plan; cre­ate a vi­able data­base; and es­tab­lish a base­line value, is sub­stan­tial.

For a small com­pany, the time re­quired to de­velop a vi­able pro­gramme will be about three man-months.

Here are 10 steps that can help you im­ple­ment a suc­cess­ful to­tal plan pre­dic­tive main­te­nance pro­gramme.

1) De­ter­mine costs

The most dif­fi­cult step in the ini­tial jus­ti­fi­ca­tion of a pre­dic­tive main­te­nance pro­gramme is the de­ter­mi­na­tion of ac­tual main­te­nance costs. Most plants do not track all con­trol­lable costs that are di­rectly driven by the main­te­nance oper­a­tion. In most cases, the costac­count­ing func­tion does not in­clude the im­pact of main­te­nance on avail­abil­ity, pro­duc­tion ca­pac­ity, op­er­at­ing costs, prod­uct qual­ity and myr­iad other fac­tors that limit plant ef­fec­tive­ness.

Your eval­u­a­tion should in­clude all main­te­nance-re­lated costs as­so­ci­ated with de­lays, re­duced ca­pac­ity oper­a­tion, over­time pre­mi­ums, and prod­uct qual­ity.

Ac­cu­racy and com­plete­ness of this data set is crit­i­cal to the long-term suc­cess of your pro­gramme. The ma­jor­ity of pro­grammes that failed in the first two years fol­low­ing im­ple­men­ta­tion can be di­rectly at­trib­uted to the lack of quan­ti­fied re­sults.

2) Se­lect sys­tems and ven­dors

A con­trib­u­tor to pro­gramme mor­tal­ity is the se­lec­tion of ei­ther the wrong pre­dic­tive tech­nolo­gies or a ven­dor that can­not pro­vide long-term pro­gramme sup­port.

A to­tal plant pre­dic­tive main­te­nance pro­gramme must use a com­bi­na­tion of mon­i­tor­ing and di­ag­nos­tic tech­niques to achieve max­i­mum ben­e­fits. None of the in­di­vid­ual tech­nolo­gies, such as ther­mal imag­ing and vi­bra­tion, pro­vide all of the ca­pa­bil­i­ties to eval­u­ate crit­i­cal plant process and sys­tems. What com­bi­na­tion of tech­nolo­gies is best for your plant?

The pre­dic­tive re­quire­ments of each plant are dif­fer­ent. As a min­i­mum, your pro­gramme should in­clude (1) key op­er­a­tions pro­cesses anal­y­sis, (2) ther­mal imag­ing, (3) process pa­ram­e­ters, and (4) vis­ual in­spec­tion.

Lu­bri­cat­ing oil and wear par­ti­cle anal­y­sis (tri­bol­ogy) should be used only where the added in­for­ma­tion de­rived will jus­tify the costs.

3) Train­ing re­quire­ments

Most pre­dic­tive main­te­nance ven­dors will of­fer some level of train­ing. How­ever, most of these train­ing pro­grammes are di­rected to­ward the use of a spe­cific sys­tem, i.e. soft­ware and in­stru­men­ta­tion, rather than com­pre­hen­sive use of the tech­nol­ogy.

There are a num­ber of ven­dors that of­fer tech­ni­cal train­ing that can sup­port your pre­dic­tive main­te­nance pro­gramme. How­ever, you should care­fully eval­u­ate the merit of their cour­ses be­fore elect­ing to use them as train­ing sup­port. In gen­eral, in­de­pen­dent train­ing com­pa­nies with no as­so­ci­a­tion with equip­ment man­u­fac­tur­ers, can pro­vide high qual­ity train­ing with an un­bi­ased ap­proach.

4) Man­age­ment sup­port

Lack of a to­tal com­mit­ment from plant or cor­po­rate man­age­ment to pro­vide the re­sources re­quired to im­ple­ment and main­tain a pro­gramme is the sin­gle largest rea­son for fail­ure of pre­dic­tive main­te­nance pro­grammes.

There are a num­ber of rea­sons for lack of long-term com­mit­ment. How­ever, in most cases, it stems from the lack of plan­ning and jus­ti­fi­ca­tion in the pre­pro­gramme ef­fort. Man­age­ment must know the true cost and po­ten­tial ben­e­fits of the pro­gramme be­fore it be­gins.

Af­ter im­ple­men­ta­tion, they must be con­tin­u­ally in­formed of the progress and ac­tual ben­e­fits that the pro­gramme pro­vides. There­fore, it is im­per­a­tive that a vi­able means of quan­ti­fy­ing the ac­tual re­sults of the pro­gramme be de­vel­oped and the on­go­ing sta­tus of the pro­gramme com­mu­ni­cated to all key man­age­ment staff.

Man­age­ment sup­port should in­clude im­ple­men­ta­tion of a for­mal main­te­nance plan­ning func­tion, a vi­able in­for­ma­tion man­age­ment pro­gramme and crafts­man skill train­ing in or­der to gain max­i­mum ben­e­fits from pre­dic­tive main­te­nance.

5) De­velop a pro­gramme plan

A def­i­nite pro­gramme plan that in­cludes all ac­tiv­i­ties re­quired by a to­tal plan pre­dic­tive main­te­nance pro­gramme must be de­vel­oped be­fore im­ple­ment­ing your pro­gramme. The pro­gramme should in­clude: spe­cific scope of pro­gramme, goals and ob­jec­tives,

meth­ods that will be used to im­ple­ment and main­tain and eval­u­ate the pro­gramme.

The plan should also in­clude spe­cific re­turn-on­in­vest­ment (ROI) mile­stones that can be used to mea­sure the suc­cess of the pro­gramme.

6) Ded­i­cated per­son­nel

The pro­gramme can­not be im­ple­mented or main­tained with part-time per­son­nel. Re­gard­less of the pre­dic­tive main­te­nance tech­niques used for the pro­gramme, reg­u­lar, pe­ri­odic mon­i­tor­ing of crit­i­cal plant pa­ram­e­ters is an ab­so­lute ne­ces­sity. Most pro­grammes im­ple­mented with part-time staff have failed be­cause ac­tiv­i­ties re­quired to main­tain the pro­gramme have been de­layed or ig­nored be­cause of other press­ing de­mands on staff time.

7) Es­tab­lish ac­count­abil­ity

Staff com­mit­ment is an ab­so­lute re­quire­ment for a suc­cess­ful pro­gramme. With­out this to­tal com­mit­ment, the pro­gramme will prob­a­bly fail. Di­vi­sion or area man­agers must ac­cept re­spon­si­bil­ity for pro­gramme suc­cess. In most plants, these man­agers con­trol the re­sources, both fi­nan­cial and per­son­nel, within their de­part­ments. With­out their full sup­port, lit­tle can be ac­com­plished.

8) De­velop a vi­able data­base

All pre­dic­tive main­te­nance tech­nolo­gies de­pend on a clear, de­tailed def­i­ni­tion of the crit­i­cal equip­ment that is in­cluded in the pro­gramme.

Data­base de­vel­op­ment re­quires a tremen­dous ef­fort in both man­power and time. A typ­i­cal mi­cro­pro­ces­sor-based pre­dic­tive main­te­nance pro­gramme may re­quire an aver­age of 1 to 3 man-years. The ini­tial in­vest­ment will greatly re­duce the man­power and time re­quired to main­tain your pro­gramme.

9) Main­tain the pro­gramme

Many pro­grammes fail be­cause the staff did not fol­low through af­ter the de­vel­op­ment stage. Fol­low the pro­gramme plan. Meet each of the sched­ules and mile­stones de­vel­oped in the pro­gramme plan. Con­stantly eval­u­ate the pro­gramme’s progress and cor­rect any er­rors or prob­lems that may ex­ist. A suc­cess­ful pre­dic­tive main­te­nance pro­gramme must be dy­namic. Fol­low through.

10) Com­mu­ni­cate

Com­mu­ni­ca­tion is ab­so­lutely nec­es­sary for long-term suc­cess. All suc­cess­ful pro­grammes have a well-de­fined com­mu­ni­ca­tions plan that in­cludes trans­mit­tal of cor­rec­tive ac­tions iden­ti­fied by the pro­gramme; feed­back from man­u­fac­tur­ing; and a reg­u­lar pro­gramme sta­tus re­port that is cir­cu­lated through­out the plant and cor­po­rate man­age­ment team. Pro­gramme jus­ti­fi­ca­tion is a never-end­ing process. Fail­ure to com­mu­ni­cate will se­verely re­duce the po­ten­tial for a suc­cess­ful pro­gramme. The Pay­off

Pre­dic­tive main­te­nance can trans­form the main­te­nance oper­a­tion from an ex­pen­sive sup­port func­tion to a full mem­ber of the profit gen­er­at­ing team. Like all things of value, a cer­tain amount of ef­fort is re­quired to gain pos­i­tive re­sults.

Source: Fluke white paper.

De­vel­op­ing and im­ple­ment­ing a suc­cesse­ful and worth­while pre­dic­tive main­te­nance pro­gramme takes time, ef­fort, re­sourses and full man­age­ment sup­port.

Newspapers in English

Newspapers from New Zealand

© PressReader. All rights reserved.