|
Root
Cause
Failure
Analysis
|
|
by Joy LePree
|
|
As a maintenance professional you probably spend a lot of time putting
out fires: administering quick fixes to small, nuisance-type problems.
It's just part of the job. |
|
Or is it? Did you know tending these chronic failures can eat up 80% of
your maintenance budget? Well, there's a way to entirely eliminate these
problems - it's called root-cause failure analysis (RCFA), and it has the
potential to save your facility millions of dollars in repair costs and
downtime. |
|
RCFA is a simple, yet disciplined process used to investigate, rectify
and eliminate equipment failure, and it's most effective when directed
at chronic breakdowns. |
|
Describe the failure
event
|
|
Describe the failure
modes
|
|
Hypothesize
|
Verify the hypotheses
|
|
Determine physical
roots & verify
|
|
Determine latent
roots & verify
|
|
|
"RCFA is applicable anywhere, but what you use it on determines how well
it affects your bottom line," says Robert Latino, vice-president of Reliability
Center, Inc., a Hopewell, VA-based reliability engineering consulting firm
that provides RCFA training to the manufacturing industry. |
|
"If you use it only in a reactive manner when you have a major, but sporadic
event, it will solve the problem," he notes. "But you will save a lot more
money if you use it on chronic failures. |
|
Latino has found that approximately 80% of a typical maintenance budget
is stored away for chronic failures, meaning that these events cost far
more, in aggregate, than major breakdowns. So it makes sense that the greatest
savings comes from applying RCFA to routine breakdowns. |
|
Rick Kalinaukas, reliability engineering supervisor for Union Camp Paper
Co., Chesapeake, VA, says his company initiated RCFA for just that reason. |
|
"We found that a large portion of our downtime came from small events that
occurred on a very frequent basis, rather than big, sporadic one-time failures,"
Kalinaukas says. "Chronic items typically slipped by our system of addressing
and prioritizing things because they just seemed to be inconveniences." |
|
Kalinaukas turned to RCFA to develop a method that addressed those chronic
failures. |
|
"The power of the process is that it shows you how to find the latent roots
responsible for the breakdown," Kalinaukas says. "Typically, most organizations
have always found the broken pieces and come up with an explanation as
to why it broke. But root-cause failure analysis takes you beyond that
to the latent roots, which are the management system weaknesses. Once you've
found these, you have the means to solve many other potential problems
that haven't yet occurred." |
|
Latino explains that in most failures there are actually three layers.
First is the physical component, then there is the human error, and finally
the latent root of the problem. The latter is always the true cause of
the problem. |
|
"Inevitably in any failure there's going to be human error, either someone
did something wrong or forgot to do something," he says. "But when you
get into true root-cause failure analysis, you get deep into management
systems. |
|
"These include training mechanisms and plant policies, procedures and specifications,"
he notes. "People make decisions based on these, and if the system is flawed,
the decision will be in error and will be the triggering mechanism that
causes mechanical failure." |
|
A prime example of these layers can be seen in a catastrophic fan failure
experienced in Chevron Corp.'s San Ramon, CA facility. Supervisor of Maintenance
and Operations Walt Flannery says following a major fan breakdown, the
maintenance department performed RCFA and determined that misshapen bearings
were the cause. Further analysis revealed that the defect was due to neglecting
to rotate the fan when it was in storage prior to installation. |
|
According to Latino, the bearing failure was the physical component, and
the lack of rotation when in storage was the human element behind it, but
the latent root cause was the facility's storage system. |
|
Corrective action came when the maintenance crew shared the information
derived from the analysis with the folks in the storage room so they would
know to have large machinery delivered on a just-in-time basis or rotated
weekly to prevent future failures. |
|
Both Flannery and Latino stress that RCFA should be used for seeking out
flaws within management systems, and not for laying blame on an individual. |
|
In order to reveal a faulty system, a multi-step RCFA process is necessary.
The method taught by RCI consists of six steps: a failure modes and effect
analysis; the preservation of failure information; the organization of
an analysis team; the actual analysis; sharing the findings and making
recommendations; and tracking the results. |
|
Of these steps, failure modes and effect analysis and the actual analysis
are the most significant. The first determines which failure events represent
your facility's most significant losses. Using this technique, notes Latino,
you generally find that 80% of a plant's losses are represented by less
than 20% of its failure events. These occurrences are referred to as the
"significant few" and are the events that should receive top priority for
RCFA. |
|
During the actual analysis, a logic tree is used to work through a failure.
To employ this tool, the failure event is placed in the top block. Under
that, all failure modes, or possible causes of breakdown, are listed. The
following layer is reserved for hypotheses of how the various failure modes
could have occurred. Next comes the verification of the hypotheses to identify
which one actually led to the problem. The next steps consist of determining
and verifying the physical roots, human roots and latent roots behind the
failure. Keep in mind that a separate logic tree must be maintained for
each analyzed event. |
|
Vernon Kingsbury, maintenance coordinator analyst with Lafarge Corp., Alpena,
MI, says his company began using RCI's RCFA process in June 1995. One of
the earliest applications of RCFA examined the frequent occurrence of a
drag conveyor failure. |
|
For the analysis, the problem was labeled "drag conveyor failure." Failure
modes were then identified -- either it was an electrical problem or a
broken link. |
|
After determining that a broken link was the cause of breakdown, Kingsbury
says the next step was to develop theories about how the link was damaged
-- either overloading the conveyor or using an inadequate drag chain. |
|
For verification, Kingsbury says the analysts examined the items hauled
by the conveyor, as well as the material of which the drag chain was constructed
of. As it turns out, the chain in use was recommended by the OEM, so the
problem was related to overloading. |
|
The next step asked "How can the conveyor be overloaded?" Several hypotheses
were studied, and the problem was soon identified as sporadic placement
of too much material on the conveyor. |
|
Kingsbury says they studied possible reasons for intermittent overloads.
Finally, a lack of control over one of the plant's processes was determined
to be the root-cause of the drag conveyor failure. The analysts examined
the process in question and made recommendations for gaining control of
it to prevent conveyor loads from overwhelming attendants in the future. |
|
"Identifying the real root-cause solved the problem with the drag conveyor
and has had a potential savings of $193,000," notes Kingsbury. He adds
that this incident was just one of many chronic failures in the facility
that have been corrected through RCFA. |
|
"Since June of 1995, I've trained 344 people in RCFA at all of our company's
different facilities," says Kingsbury. "We normally expect to save a little
over $2.25 million over a two year period for every 100 people trained."
The potential savings for the company are expected to be around $13 million. |
|
The savings realized by Chevron are also impressive. In the fan scenario
alone, the company saved a fortune. Because the fan that originally broke
down was one of eight kept in storage prior to placement in the facility,
maintenance crews decided to run vibration tests on the other seven machines
to determine if they also contained faulty bearings. The tests detected
a common vibration signature that indicated possible failure of all seven
in the future. So the equipment was torn down and the bearings were replaced
and realigned. |
|
"Performing those activities cost $2,300," says Flannery. "We spent $32,000
to remedy just one catastrophic fan failure, so we saved $224,000 in repair
costs alone." This figure does not include the potential savings from eliminating
downtime or from educating employees about proper storage techniques. |
|
Just one RCFA analysis in Kalinaukas' Union Camp paper mill, saved the
company $1,021,000 in production losses. The problem was a packing failure
on a large piece of machinery that resulted in frequent in downtime. Originally,
he says, the mechanics responded to the frequent breakdown in a reactive
way by replacing the part. Eventually, a detailed root-cause failure analysis
identified the problem: the wrong packing sleeve was placed in the machine
during an earlier repair due to incorrect purchasing procedures. |
|
To solve the problem and realize the impressive savings, the packing sleeve
was replaced with the right one, and a new purchasing system was initiated
to insure that the wrong part will never be ordered again. |
|
"The paper industry is such a capital intensive industry, and our throughput
rates are so phenomenally large that any small component failure causes
tremendous profit loss," notes Kalinaukas. "That's what makes root-cause
failure analysis an excellent process for our mill." |
|
Frank Meitz of the Bonita Springs, FL-based Frank J. Meitz Institute of
Maintenance Management, a division of DP Solutions, Greensboro, NC, says
process industries probably get the highest return immediately because
they have a tremendous amount of bearings, drives, motors, and equipment
powered by different applications than most other facilities. |
|
However, the potential savings are substantial for any industry because
RCFA promotes reduction of chronic problems and increased mean time between
failures in all situations. Other benefits derived from the RCFA process
include improved product quality and regulatory compliance through increased
reliability, better sensitivity to machinery problems, less non-value-added
work and administrative obstacles, and fulfillment of the ISO 9000 certification
requirement for "corrective action," according to the Reliability Center. |
|
But what makes RCFA especially appealing is that the potential benefits
far outweigh any capital investment in the process. "This is really a very
inexpensive way to save your company a lot of money and time," says Kingsbury.
"When we began using root-cause failure analysis all that we really paid
for was the training and the text we use." |
|
The training course offered by RCI runs about $1,000 and the text is about
$200, according to Kingsbury. He says one trained person in the facility
can pass the knowledge and methods onto the rest of the plant. |
|
Lafarge's training program is set up very simply. A group is gathered off-site
for a one day session and Kingsbury teaches them RCFA methodology. About
four to six weeks later there is a reunion where the students take an actual
problem they've encountered in the facility and perform RCFA. At that time
the group graduates and the results of the analysis are used to implement
any necessary changes in management systems. If it sounds too easy to be
effective, consider the fact that the drag conveyor failure was remedied
in one of the earliest training sessions. |
|
After training, most RCFA users find that they can get by without purchasing
a lot of new tools or technology. "It has really been a relatively no cost
process for us," says Kalinaukas. "We haven't had to purchase any additional
equipment because we can use equipment that's already in place, like vibration
monitoring." |
|
However, it should be noted that unless your facility has an in-house lab
for things like metallurgical examinations, you may have to turn to an
outside source for some analyses. "Depending on a plant's in-house capability
to perform internal measuring and analysis, support from an external service
may be needed," says Meitz. Finite element analysis, physical modeling,
rotor dynamic and metallurgical analysis, says Latino, are the most common
tests required for RCFA. |
|
"In some component failure situations, we'll have items professionally
analyzed against original design," says Kalinaukas. "We might send out
a sample of packing sleeve material and ask for a professional opinion
as to how that component failed. |
|
"We then receive a formal report stating that a bearing failed due to a
lack of lubrication and then we can continue with our RCFA process," he
says. "For the most part we use external expertise to back up our conclusions." |
|
Kingsbury's Lafarge also employs external sources. "We do use outside services
to identify some problems, but not that often because not every theory
needs outside expertise to be proven," he notes. "In most cases hypotheses
can be verified in-house with the tools already available." |
|
Chevron, says Flannery, is lucky to have its own in-house research and
technology lab for destructive and non-destructive failure testing. When
the fan failed, bearings were sent to the lab for metallurgical testing. |
|
Despite the cost of outside services, there will still be a worthwhile
return on investment. "You should expect three things to happen," says
Meitz. "Overall maintenance costs should be reduced because the maintenance
staff will be able to do more work through proper planning and scheduling,
rather than reactive maintenance." |
|
He continues: "Based upon the institution of RCFA, investment in spare
parts should be reduced, which will bring a recurring savings on carrying
costs. Still, the biggest savings will come from increased use of assets.
These three occurrences should really support the input of RCFA." |
Use RCFA to investigate
and
eliminate equipment problems.
It's most effective when
directed
at chronic failures.
|
|
|
With all the benefits of RCFA - namely putting an end to chronic failures
and nuisance jobs and saving millions of dollars in maintenance costs -
is there really any reason not to consider initiating RCFA in your plant? |
RCI Offers the full
range of Reliability Consulting Services and Training Programs for Industry.
We conduct facilitations, reliability assessments, FMEA & Root Cause
Failure Analysis Training - Public & On-Site.
For more information
contact:
Reliability Center, Inc.
P.O. Box 1421
Hopewell, Virginia 23860
Phone: (804) 458-0645
Fax: (804) 452-2119
Website: http://www.reliability.com
|
|
Return
to Failure Analysis Reference Library Index
|
