Introduction

After you buy a new car or house when one thing breaks or scratches you would not buy a new one, isn't it? The best way is to maintain it so that you can use it for a more extended period. With the evolution of technology, the development of the wireless network, highly accurate sensors and powerful software analytics set the stage for the world to start using advanced proactive maintenance techniques such as condition based maintenance for monitoring complex assets. Markets and Markets predicted that the conditioning monitoring market might grow from $2.38 billion in 2018 to $3.5 billion by 2024.

Condition Based Maintenance

Now let us get started with the following questions:

What is condition-based maintenance?

CBM or Condition Based
Maintenance is maintenance in which maintenance operations perform depending on
the conditions of your assets at present.

The asset’s condition is
checked through visual inspections, tests and performance data that are often
gathered by different sensors or tools or both.

The data above shows that you
can schedule maintenance work just before a particular piece of equipment might
fail.

To explain this, you can install
a sensor that can measure the vibrations of a particular piece of rotating
equipment, and over time the moving work will start degrading and fall out that
might increase the amount of vibration. When you install the sensor, they can
warn you when the amount of vibration crosses the set limit and remind you
about replacing it shortly.

Differences between condition based maintenance and predictive
maintenance

It is not right for us CBM and predictive maintenance interchangeably, and there is a lot of overlap between condition based maintenance and predictive care but not the same!

Predictive maintenance has a combination of
condition-based diagnostics with complex predictive formulas to get the exact
prediction of the piece of equipment falling. The condition-based diagnostics includes
measuring vibrations, temperature and other variables having a predictable
pattern.

On the other side, CBM depends on set intervals
and lacks the predictive formulas used to interpret different trends.

Thus, predictive maintenance is a more accurate
version than the condition-based maintenance.

The goal of condition based maintenance is helping one optimize the maintenance resources by performing maintenance work only when needed.

Condition Measurement vs Condition Monitoring

Operators should know the status of a machine at any point
in time to perform maintenance proactively. Therefore, operators need to
understand what is going on and then use the available information for
determining when to intervene.

To achieve this objective
condition monitoring makes it straightforward. The process of establishing the
condition of an asset by monitoring specific operating parameters of the
equipment is condition monitoring. Condition monitoring aims to identify
significant changes or deviations that indicate a fault developing within. The
primary support for predictive maintenance is condition monitoring.

For carrying out such
monitoring, it is necessary to take measurements at regular intervals, and the
measures are called condition measurement, which is essential for determining
the asset's condition getting monitored. The collection and analysis of these
measurements reveal the results and enables accurate diagnosis of the
equipment's state. For prevention of failure and ensuring continuous
availability of the equipment, the maintenance team plays the required
maintenance actions.

Vibrations, temperature,
pressure, oil and noise are some of the common conditions to measure. Without
condition measurement, there cannot be condition monitoring.

Techniques in condition monitoring

Non-invasive data collection activities let
obtain the measurements that are the basis for monitoring and equipment's
operating conditions. Using sensors, transducers and different condition
monitoring tools which are portable instruments, the condition measurements can
be collected in a non-destructive manner either continuously or at intervals.

Usual condition monitoring techniques used are:

Vibration
Analysis

The commonly used monitoring technique for
rotating equipment like compressors, motors, centrifugal pumps is vibration
analysis. The vibration sensors, when installed monitor axial, vertical or
horizontal movements and send notifications when it exceeds the limit.

Lubricant
analysis

Another non-invasive technique that can exhibit
the internal condition of a machine is oil or lubricant analysis. By analyzing
the size and number of particles or debris like iron, aluminium silicate,
silicon, etc. is works by collecting oil samples to determine asset wear.

Infrared
thermography

The temperature increases when the object's
amount of radiation emitted is increased. These are quickly detected by
infrared cameras as these are invisible to the human eye. The cameras check the
temperature irregularities regularly in energized equipment.

Ultrasound
Testing

Ultrasonic sensors can detect sound generated by a piece of malfunctioning equipment. This ultrasound technique applies to a wide range of machinery from both high-speed and low-speed equipment to high-pressure fluid situations.  There are issues like deep subsurface defects, leaking gases and over or under lubricated bearings that alerts the operators. The technique is ideal for specific electrical inspections like closed gears for safety reasons. Besides these, one can use corrosion monitoring, current motor analysis, visual inspection techniques also.

When and How to Apply CRM?

Condition monitoring is
usually implemented in a hastened way without a thorough understanding of a few
factors that will affect its effectiveness in the overall system and condition
monitoring plays a beneficial role in the maintenance program.

Consider the following
factors:

Does the solution monitor
specific failure modes?

A definite cause of the
failure or one of the known ways in which a system can fail is a failure mode
and the more complicated equipment, the more failure modes it can have. The
impact and understanding of these failure modes help to identify and adopt the
right condition monitoring solution, and this is a vital aspect of improving
asset reliability.

Does the technology interface with existing ERP or CMMS
systems?

Necessarily all the data
channels can interface effectively with minimal or zero disruptions to
operations with the continuous advancements the way we use data.

Computerized Maintenance
Management System (CMMS) and Software like Enterprise Resource Planning (ERP)
channels would be included. Wireless technology from the Internet of Things may
also have incorporated tools by some organizations.

What is the ease of implementation?

Some condition solutions and
monitoring sensors are more accessible to implement than others. One needs to
ultimately scope and understand the resources necessary to implement a specific
condition monitoring solution. Time, the effort required, and labour for
implementation should be considered, and these depend on how large the system
is, how complex the machines that are used, how easy it is to install the
sensors needed and how fast can your team adopt the new solutions.

Can alerts be generated accurately and improved over time?

Do consider how reliable the
generated alerts will be before applying a particular solution.

A reason why an organization
may decide to abandon the process after implementation is because of false
positives. It is not uncommon in condition monitoring settings. Everyone will
lose confidence in the system if technicians are continuously called on to
respond to failure that ends up being false many times. Thus, from your
proposed solution provider, check the record of the ratio of actual versus
false positives.

The P-F interval and the P-F curve

A method for explaining how to detect failure
is the P-F interval and P-F curve. The P-F curve illustrates the behaviour of a
machine as it approaches functional loss.

When failure begins to manifests, the curve
shows the deteriorating of the equipment gradually to a point where it can be
detected, and this pint is called the potential failure. Functional or hard
loss is the end of failure is not detected and mitigated, and the situation
continues until the system fails.

The interim between P and F must be sufficient
for analysis to be useful and corrective action taken in condition monitoring
as this whole process would become pointless. One crucial point is that the
monitoring time should be shorter than the span of the P-F interval. Within the
P-F interval, using multiple monitoring points is better and advisable. 

To establish a condition based maintenance program here are the following steps

1. The assets you want to monitor
   select them as this is the case with the preventive maintenance plan.
   Concentrate on investments that are important to production or organizational
   success, costly to repair and replace, and not going to be soon replaced.
2. The best way to identify all known
   and possible failure modes is by performing an RCM analysis and focus on the
   failure modes, which are manageable using CBM strategies.
3. You need to select accurate
   CBM solutions and monitoring techniques for the failure modes to identify.
4. Baseline limits need to be
   defined to get a warning by the system about the system starting to
   deteriorate, and corrective actions have to be performed.
5. Collecting and recording
   measurements to establish the CBM program should be in the centre of the plan.
6. Analyze the data coming from
   inspections and sensors to plot a trend and schedule maintenance work
   accordingly.

Benefits of Condition Based Maintenance

CBM is not an exception, as it
is a proactive maintenance strategy and bring a similar advantage to the table.

Some benefits of using condition based maintenance are:

• Reduction in the number of
  unplanned failures
• Availability, reliability and
  worker safety improved equipment.
• Time spent on maintenance is
  minimized by only doing maintenance work when needed.
• During non-peak times repairs
  can be scheduled.
• Increase in the lifetime of
  the asset
• Improvement in the performance
  equipment
• Reduction and minimizing
  inventory costs like ordering a part when you plan to repair and not to hold as
  much inventory

These
all mentioned above are not free of charges, and some do come with a price

What are the Challenges of condition based maintenance?

Some challenges faced by CBM
are:

• It is expensive to install
  condition monitoring tools.
• It is not always
  straightforward about choosing the right sensor.
• Considerable time and money
  need to be spent to train the employees for using CBM effectively.
• In harsh operating conditions,
  sensors might have trouble working correctly.
• The sensors can get damaged
  when working in harsh operating conditions and may lead to replacing them
  regularly, which is not cheap.
• Unpredictable peak time may
  occur as maintenance is done when data shows it is needed. Changes of multiple
  assets requiring care may show up.

Requirements for the implementation of condition based maintenance

The following requirements are
needed for the implementation of CBM:

• Condition monitoring sensors
• Condition monitoring tools
• Trained staff for effective
  use of CBM technology
• Maintenance manager or whoever
  knows to analyze the data and schedule work accordingly
• A CMMS solution
• A willing maintenance team to
  learn and adapt to changes in the workflow

Conclusion

Condition based maintenance needs to be decided by organizations on a case to case basis. The only way to stay competitive in any production environment is reaping the benefits of proactive maintenance.