| Reliability-centered maintenance (RCM) continues
to show up prominently in technical literature as the future strategic
direction in machinery maintenance. It should be. RCM is the right thing
to do when it comes to optimizing the operational reliability of plant
equipment. It is important for lubrication engineers, oil analysts and
other tribology professionals to understand RCM and how oil analysis and
lubrication management fit into the RCM picture. |
| RCM is the systematic process with which to
optimize reliability and associated maintenance tactics with respect to
operational requirements. Economic optimization of machine reliability
relative to organizational goals is the primary objective of the RCM process.
Simply stated, RCM helps us ensure that if we spend a dollar on improving
reliability, we are getting the full dollar back, plus some acceptable
return on the investment. |
| As Figure 1 illustrates, the law of diminishing
marginal returns applies to the implementation of reliability improvement
measures. Generally speaking, the first dollar invested in reliability
improvement tends to yield a higher return-on-investment than any dollar
subsequently invested. The objective is to reach the point of optimization
at which the benefits of reliability, expressed as total operating costs,
are maximized through cost reduction. RCM is a set of systematic engineering
procedures for achieving and maintaining that objective. |
| RCM's roots trace back to the 1960s when it
was advanced to improve the safety and reliability of commercial aircraft.
Since then, it has begun to move into the industrial sector as a result
of work conducted by several authors, most notably John Moubray (RCM II)
and NASA's publication "Reliability-Centered Maintenance Guide for Facilities
and Collateral Equipment". Going further back, however, RCM owes its origins
to the development of the reliability engineering discipline. It was there
that the fundamental analytical tools were created to estimate the reliability
of electrical and mechanical components and systems. Simply stated, RCM
is a component of the quality movement that is focused on improving the
safety, reliability and productivity of the equipment upon which our society
depends for transportation, power and energy, and goods and services. |
| Why
RCM and Why Now? |
| In North America, we simply are not building
new factories. In an economy where prices are set globally, we must profitably
produce products (polypropylene to Plymouth Vipers) with aging equipment
operated and maintained by a workforce that is among the most expensive
in the world. This means that manufacturing assets must deliver big. And,
so too must the maintenance strategies, like RCM, to maximize profitability. |
|
|
Figure 1
|
| RCM guides the reliability investment with
improvement measures and techniques including lubrication management and
analysis such that the economic optimization is realized. NASA has identified
specific guiding principles of RCM (see sidebar). But, essentially, the
reliability engineer is tasked to answer the following questions: |
-
What is the system or equipment
asked to do?
-
What functional failures
are likely to occur?
-
What are likely consequences
of these functional failures?
-
What can be done to prevent
these functional failures?
|
| In the past, we attempted to achieve reliability
with frequent rebuilds. The strategy was founded on the assumption that
the failure rate of machines increased as the asset aged. While some items
fail in this manner, most complex systems, e.g., those found in process
and manufacturing plants, do not. In one study; 30 identical deep groove
ball bearings were run to failure on a test stand under highly controlled
conditions. The variation in failure times was so great that if one statistically
estimated the appropriate replacement time at the 95% confidence level
the machine would never be started! In the field, the variation in time-to-failure
is even greater. Thus, we have learned that the time at which complex equipment
should be rebuilt can not, in many cases, be effectively estimated. |
NASA's
Guiding Principles of RCM
| RCM is Function Oriented - It seeks to preserve system
function, not just operability for operability's sake.
? RCM is System Focused - It is more concerned
with systems than components.
? RCM is Reliability-Centered - It treats failure
statistics in an actuarial manner. It seeks to define the probability that
a system can perform its intended function for a specified operating interval
under stated conditions.
? RCM Acknowledges Design Limitations - It seeks
to maintain inherent reliability with the understanding that changes in
inherent reliability is the province of design rather than maintenance.
It also recognizes that input from the maintenance organization is critical
to continuous improvement of system design.
? RCM is Driven by Safety and Economics -Safety
(human and environment) must be ensured at any cost. Thereafter, cost-effectiveness
is the evaluative criteria.
? RCM Defines Failure as Any Unsatisfactory Condition
- Therefore, a failure can be defined as loss of function (operation ceases
or falls below minimum capacity requirements), or a loss of acceptable
quality (operation continues).
? RCM Uses a Logic Tree to Screen Maintenance Tasks
- This provides a consistent approach to maintenance of all kinds of equipment.
? RCM Tasks Must Be Effective - The techniques
must be technically sound and cost effective.
? RCM Tasks Must Be Applicable - The tasks must
reduce the number of and/or the impact of failures.
? RCM Acknowledges Four Types of
Run-to Failure
Time-Directed Maintenance
Condition-Based Maintenance
Failure-Finding/Proactive Maintenance
? RCM is a Living System - It maintains a feedback
loop with which to facilitate continuous
|
|
| More recently, we have employed vibration
analysis, lubrication analysis, thermography, and other condition monitoring
and predictive maintenance tools in an attempt to identify early stage
failures so corrective action can be scheduled "on-condition". We have
also applied proactive measures to monitor and control the root causes
of degradation and failure such as lubricant contamination, wrong/degraded
lubricant, misalignment, unbalance, etc. These measures that employ advanced
maintenance techniques and technologies have proven very effective, but
if over-applied, they can be expensive and counterproductive. Moreover,
in some cases they simply don't provide the required improvement in reliability
to get the job done. In these instances, system redesign or the employment
of redundancy is required to achieve the goals of the organization. |
|
|
Figure 2 - RCM Flow Diagram
|
| The process by which a reliability strategy
is selected according to RCM is very systematic and logical (see Figure
2). As the flow diagram suggests, assets are audited with respect to their
role in over-all system reliability and productivity. If acceptable, no
changes are required. If unacceptable, then questions about the criticality
of the asset define the need to identify the most efficient means of attaining
the necessary reliability. If the asset is deemed non-critical, for example,
it is simply run to failure then rebuilt or replaced. For mission critical
systems, advanced maintenance techniques are typically the first choice
because their use is relatively inexpensive compared to redesign and the
employment of redundancy. |
| In some cases, redesign or employment of redundancies
is required to meet the objectives of the organization. Redesign in the
form of proactive measures to control (and monitor) lubricant contamination,
alignment, balance, etc. is usually much less expensive than to deploy
than failure detection strategies. Conversely, more involved system redesign
is usually very expensive and often produces unpredictable results. The
employment of redundant systems is the most expensive method to improve
reliability, but it provides very sure results. Employment of RCM helps
to avoid the casual application of the latest "panacea" strategy, avoiding
mistakes that waste resources and provide mediocre and unpredictable performance. |
| Table 1 summarizes strategies for achieving
reliability and the conditions under which they are selected in the RCM
process. In today's competitive environment, organizations are looking
to advanced maintenance strategies, especially condition-based maintenance,
to provide the necessary reliability at minimum cost. The cost to rebuild
or replace is quite high and yields dubious value. Purchasing and maintaining
redundant systems is reserved for only the most critical systems where
no other strategy provides satisfactory results. |
|
Maintenance Strategy
|
Action Required
|
RCM-Based Application
|
| Run to failure (reactive) |
Repair or replace upon failure |
Non-critical. Costs to control
or detect failure exceeds benefits. |
| Scheduled discard to restoration
(preventive) |
Repair or replace on time
or cycles |
Asset has a well documented
MTBF and a small standard deviation |
| On-condition maintenance
(predictive) |
Employ condition monitoring
to detect early stage failures. Replacement or repair are scheduled on-condition. |
Asset fails randomly. Critical
nature justifies early detection techniques. |
| Redesign and condition-control
(proactive) |
Changes in hardware, loading
or procedures. Condition monitoring detects the presence of root causes
of failure. |
Objective is to reduce the
failure rate for a given time period |
| Redundancy |
Deploy active shared-load
or stand-by redundant systems |
Mission critical assets
for which no other approach is acceptable |
|
|
Table 1
|
| Advancing technology has brought condition-based
maintenance to the forefront of the RCM movement. Lubrication management
and oil analysis play an integral role in this movement. |
|
|
Figure 3
|
| The reliability engineer employs a number
of analytical tools to optimize reliability relative to mission goals.
Some of the more common tools include: |
-
Reliability Statistics - Reliability statistics differ from conventional
experimental statistics. They provide the means with which to estimate
the likelihood that a system will achieve its mission given a stated duration
and operating conditions. It is important to become knowledgeable about
the methods of reliability engineering in advance of undertaking an RCM
project.
|
-
Reliability Block-Diagrams - Once sub-system reliability is determined,
the system can be effectively modeled from the reliability perspective.
Once modeled, the weak links usually become evident and can be addressed
with reliability growth measures to eliminate the deficiencies. Figure
3 illustrates block-diagrammed examples of simple serial, parallel and
combination systems.
|
-
Failure Modes Effects and Criticality Analysis (FMECA) - FMECA is
the inductive process of identifying Primary functional failures, their
related failure modes or states, the effect of the failure modes on the
operation of the system and the associated criticality of the failure mode
as a function of impact and likelihood. This valuable analytical tool enables
the removal or better management of failure modes through application of
advanced maintenance techniques, redesign or redundancy.
|
-
Root Cause Failure Analysis (RCFA) - RCFA assesses a failure after
the fact with the intent to determine its root cause for occurrence. Once
the root cause is ascertained, the engineer can assess the risk of recurrence,
the success with which the root cause might be controlled and the cost
to control it. With this information, a decision can be made to deploy
control measures or to let it go.
|
| RCM
and the Oil Analysis Professional. |
| After careful analysis, reliability optimization
in a process or batch manufacturing plant usually includes a heavy dose
of proactive and predictive maintenance. Typically, lubrication management
is a top candidate for improvement in the quest to bolster mechanical system
reliability. As such, the lubrication engineer or oil analysis technician
will need to provide some technical precision in the following areas: |
-
Lubricant Specific FMECA
-
Deployment of Proactive Lubrication Management Measures
-
Effective Utilization of Predictive Oil Analysis Techniques
|
| Lubricant related failures are often lumped
together somewhat casually under the term inadequate lubrication". The
lubrication engineer knows that inadequate lubrication can refer to insufficient
quantity of oil, wrong oil, degraded oil, contaminated oil, additive depletion,
poor specification or many other conditions. He must support the RCM process
with a more detailed lubrication-related FMECA that properly represents
the equipment, the operation, the environment, etc. Figure 4 identifies
several of the lubrication related failure modes and the general questions
for which the lubrication engineer should supply a FMECA information. Many
other machine specific failure modes are revealed effectively by oil and
wear debris analysis. That information must be included with technical
precision into the overall FMECA process. |
Some
Lubricant Related Failure Modes
|
Degradation Particle
contamination / Moisture contamination
|
Sudden volumetric
loss / Low levels / Wrong lube
|
Fuel/chemical dilution
/ Coolant contamination / Additive depletion
|
Foaming / .Under/over
greasing etc.
|
|
Lube Related Failure
Mode
|
Possible Effects
|
Criticality
|
Possible Causes
|
Detection Mechanisms
|
P-F Interval
|
| What is the nature of the
failure? |
What might happen if it
occurs? |
What is the importance and
likelihood of occurrence? |
What root causes led to
the failure? |
How can the failure be detected? |
How much warning is provided
by detection mechanisms? |
|
|
Figure 4
|
| Proactive lubrication management offers an
inexpensive way to reduce the inherent failure rate of mechanical systems.
When the failure rate is reduced, reliability increases for all mission
duration periods. Often, lubrication management can eliminate the need
for more drastic and expensive measures. The lubrication engineer or oil
analysis technician should coordinate with the reliability engineer to
understand which systems require reliability growth. The process should
culminate in the form of a list of changes to upgrade lube specifications,
staff education and development, improve contamination control improvements,
improve delivery mechanisms, enhance testing and inspection practices.
educate and train staff, etc. |
Oil analysis has proved to be a very effective
method for scheduling on-condition oil changes. Perhaps more important
is the effectiveness with which oil analysis can identify machine
failures and support the process of identifying the root cause of the
failure. Just as blood carries clues about the health of the human body,
oil carries important information about the health of machinery. In some
cases, oil analysis provides the very earliest warning of trouble. In other
cases, it provides confirming information. And, occasionally, it carries
no information at all about a failure. just as the physician employs all
the techniques and specialists available to detect and understand problems
related to health, the machinery engineer must select the right mix of
analysis techniques and technologies to make the very best decision. |
| The warning time in advance of a functional
failure that a monitoring technique provides is called the P-F interval.
P refers to the time at which a potential failure is detected, and F refers
to the time at which the actual failure occurs (see Figure 5). Simply stated:
the longer the P-F interval, the more time one has to make a good decision
and plan actions. As a rule, better decisions and more planning time minimize
the financial impact of the event on the organization. Table 2 summarizes
a general assessment of the effectiveness of the primary condition monitoring
tools (lube analysis, vibration analysis and thermal analysis) with respect
to the detection P-F interval and root cause failure analysis. It is always
impor-tant to factor in application and environmental conditions in such
generalizations before finalizing technology selection and deployment selection
decisions. |
|
|
Figure 5
|
| In conclusion, all reliability improving techniques
including lubricant management and oil analysis, must harmonize and align
with the organizational objective of optimized asset utilization and maximized
profit. RCM is the heart and soul of this process. The lubrication engineer
and oil analyst of the future will play a vital role in the RCM process
and as a part of the reliability team achieving and maintaining optimized
asset reliability. |
|
Detection P-F Internal
(what's going to happen)
|
Root Cause Failure
Analysis
(why did it happen?)
|
|
Lube
Analysis
|
Vibe
Analysis
|
Therm
Analysis
|
Lube
Analysis
|
Vibe
Analysis
|
Therm
Analysis
|
| Root Causes Control |
|
|
|
|
|
|
| Lubricant contamination |
excellent
|
poor
|
poor
|
excellent
|
poor
|
fair
|
| Misalignment |
fair
|
excellent
|
fair
|
fair
|
excellent
|
fair
|
| Imbalance |
fair
|
excellent
|
fair
|
poor
|
excellent
|
fair
|
| Wrong lubricant |
excellent
|
poor
|
poor
|
excellent
|
poor
|
poor
|
| Degraded lubricant |
excellent
|
poor
|
poor
|
excellent
|
poor
|
poor
|
| High operating |
fair
|
fair
|
excellent
|
fair
|
fair
|
excellent
|
| Failure Detection |
|
|
|
|
|
|
| Wear |
excellent
|
good
|
fair
|
excellent
|
fair
|
fair
|
| Cavitation |
good
|
poor
|
fair
|
fair
|
poor
|
fair
|
| Gear tooth fracture |
poor
|
excellent
|
poor
|
fair
|
fair
|
poor
|
| Structural resonance |
poor
|
excellent
|
poor
|
poor
|
excellent
|
poor
|
| Fatigue |
excellent
|
good
|
good
|
fair
|
fair
|
excellent
|
|
|
Table 2
|
