Driving Predictive

Maintenance in a
Connected World

Maintenance is a strategic concern when developing and

manufacturing a product – and for good reason.
 INDEX

 Introduction
 Detailed Study of Predictive Maintenance
 Implementing Best Practice
 Case Study
 Advantages of Digital Transformation
 Challenges
 Towards Digital Reliability
 Conclusion
 Reference
INTRODUCTION
in revenues and productivity. For example some of the leading automotive manufacturers
Predictive Maintenance in the industrial world Powered by sensors, connectivity and smart
machines, the Internet of Things (IoT) is reshaping the manufacturing and industrial processes,
effectively changing the paradigm from one of ‘repair and replace’ to more of ‘predict and
prevent’. In an industrial scenario, unplanned equipment downtime can be extremely costly to
business. Today, manufacturers and organizations in other asset heavy verticals cannot afford
to wait till a machine or equipment breaks down in order to figure out what went wrong. On
the other hand, enterprises also don’t want to spend costly time and resources doing unwanted
maintenance to all of their equipment and machinery without really needing to do so.
Predictive Maintenance, in a nutshell, is all about figuring out when an asset should be
maintained, and what specific maintenance activities need to be performed, based on an
asset’s actual condition or state, rather than on a fixed schedule, so that you can maximize
uptime and productivity. It is all about predicting & preventing failures and performing the right
maintenance routines in order to reduce costly equipment downtimes. So why is Predictive
Maintenance gaining increasing importance today, especially among asset intensive verticals? It
is driven primarily by three key factors:

 Unplanned outages equals lost revenues – For asset heavy industries, unplanned
equipment outages can mean big losses estimate that every minute of unplanned downtime
can cost them as much as $15,000 - $20,000 and a single downtime event can cost
approximately $2 Million. Given the business impact, it is not surprising that these industries
have been focusing on driving predictive maintenance to minimize downtime and losses
 Unnecessary costs and work – It costs valuable resources, time, and material to do
ongoing maintenance activities and any unnecessary maintenance done leads to unwanted
costs and redundant work. Leading organizations want to ensure that they are doing the right
level of maintenance, and replacing the right parts for the right machinery at the right time.

 Moving from reactive to proactive – Currently, most organizations employ a

combination of corrective and preventative maintenance on their machinery to minimize the
impact of unplanned downtime. However most of the current methodologies are still reactive
and after-the-fact, or leads to redundant and unnecessary work. With predictive maintenance
organizations are moving towards effectively predicting and preventing issues before they

Industrial and processing plants typically use two types of maintenance management, either
run-to-failure or preventive maintenance.
 Fig 1. Six dimension Of Maintenance and Failure

Run-to-failure management

The logic of run-to-failure management is simple and straightforward. When a machine breaks
down ... fix it. This “If it ain't broke, don’t fix it” method of maintaining plant machinery has
been a major part of plant maintenance operations since the first manufacturing plant was built
and on the surface sounds reasonable. A plant using run-to-failure management does not spend
any money on maintenance until a machine or system fails to operate. Run-to-failure is a
reactive management technique that waits for machine or equipment failure before any
maintenance action is taken. It is in true a no maintenance approach of management. It is also
the most expensive method of maintenance management.

The major expenses associated with this type of maintenance management are:
• High spare parts inventory cost;
• High overtime labor costs;
• High machine downtime and
• Low production availability.

Analysis of maintenance costs indicate that a repair performed in the reactive or run-to-failure
mode will average about three times higher than the same repair made within a scheduled or
preventive mode.

This reactive method of management forces the maintenance department to maintain

extensive spare parts inventories that include spare machines or at least all major components
for all critical equipment in the plant. The alternative is to rely on equipment vendors that can
provide immediate delivery of all required spare parts. Even if the latter is possible, premiums
for expedited delivery substantially increase the costs of repair parts and downtime required to
correct machine failures.

For some of the leading automotive manufacturers, unplanned downtime can

cost them as much as $15,000 - $20,000/ minute and a single downtime event
can cost approximately $2 Million

Preventive maintenance

There are many definitions of preventive maintenance, but all these management programs are
time-driven. In other words, maintenance tasks are based on elapsed time or hours of
operation that are based on statistical or historical data for specific types of plant equipment.
Figure 1.1 illutrates an example of the statistical life of a machine-train. The mean-time-to-
failure (MTTF) or bathtub curve indicates that a new machine has a high probability of failure
during the first few hours or weeks of operation, usually caused by manufacturing or
installation problems. Following this initial period, the probability of failure is relatively low for
an extended period of time. Following this normal machine life period, the probability of failure
increases sharply with elapsed time or hours of operation. In preventive maintenance
management, machine inspections, lubrication, repairs or rebuilds are scheduled based on the
MTTF statistic.

Please add diagram from fluke and delottie of different type of maintaince

The actual implementation of preventive maintenance varies greatly. Some programs are
extremely limited and consist of lubrication and minor adjustments. More comprehensive
preventive maintenance programs schedule repairs, lubrication, adjustments and machine
rebuilds for all critical machinery in the plant. The common denominator for all of these
preventive maintenance programs is the scheduling guideline. All preventive maintenance
management programs assume that machines will degrade within the statistical time frame
typical for its particular classification. For example, a single-stage, horizontal split-case
centrifugal pump will normally run 18 months before its wear parts should be replaced. Using
preventive management techniques, the pump would be removed from service and rebuilt
after 17 months of operation.

The problem with this approach is that the mode of operation and system or plant specific
variables directly affect the normal operating life of machinery. The meantime-between-failures
(MTBF) will not be the same for a pump that is handling water and one handling abrasive
slurries. The normal result of using MTBF statistics to schedule maintenance is either
unnecessary repairs or catastrophic failure. In the example, the pump may not need to be
rebuilt after 17 months. Therefore the labor and material used to make the repair was wasted.
The second option using preventive maintenance is even more costly. If the pump fails before
17 months, we are forced to repair using run-to-failure techniques. Analysis of maintenance
costs have shown that a repair made in a reactive, i.e. after failure, mode will normally be three
times greater than the same repair made on a scheduled basis.

“Better predictive maintenance using IoT can reduce

equipment downtime by up to 50 percent and reduce
equipment capital investment by 3 to 5 percent…In
manufacturing, these savings have a potential economic
impact of nearly $630 billion per year in 2025.”
Manyika, James, et al., “Unlocking the potential of the Internet of Things.”

Predictive maintenance

Predictive maintenance is a condition-driven preventive maintenance program. Instead of

relying on industrial or in-plant average-life statistics, i.e. mean-time-to-failure, to schedule
maintenance activities of the, predictive maintenance uses direct monitoring operating
condition, efficiency, heat distribution and other indicators to determine the actual mean-time-
to-failure loss of efficiency that would be detrimental to plant operations for all critical systems
in the plant or facility. At best, traditional time-driven methods provide a guideline to normal
machine-train life spans. The final decision, in preventive or run-to-failure programs, on repair
or rebuild schedules must be made on the bases of intuition and the personal experience of the
maintenance manager.

The addition of a comprehensive predictive maintenance program can and will provide factual
data on the actual operating condition of critical assets, including their efficiency, as well as the
actual mechanical condition of each machine train and the operating efficiency of each process
system. Instead of relying on industrial or inplant average-life statistics, i.e. mean-time-to-
failure, to schedule maintenance activities, predictive maintenance uses direct monitoring of
the mechanical condition, system efficiency and other indicators to determine the actual mean-
time-to-failure or loss of efficiency for each machine-train and system in the plant. This data
provides maintenance management the factual data needed for effective planning and
scheduling maintenance activities.

Predictive maintenance is much more. It is the means of improving productivity, product quality
and overall effectiveness of our manufacturing and production plants. Predictive maintenance
is not vibration monitoring or thermal imaging or lubricating oil analysis or any of the other
nondestructive testing techniques that are being marketed as predictive maintenance
tools.Rather, it is a philosophy or attitude that simply stated uses the actual operating condition
of plant equipment and systems to optimize total plant operation. A comprehensive predictive
maintenance management program utilizes a combination of the most cost-effective tools, i.e.
thermal imaging, vibration monitoring, tribology and other nondestructive testing methods, to
obtain the actual operating condition of critical plant systems and based on this factual data
schedules all maintenance activities on an as-needed basis. Including predictive maintenance in
a comprehensive maintenance management program will provide the ability to optimize the
availability of process machinery and greatly reduce the cost of maintenance. It will also
provide the means to improve product quality,productivity and profitability.A predictive
maintenance program can minimize eunscheduled breakdowns of all electrical and mechanical
equipment in the plant and ensure that repaired equipment is in acceptable condition. The
program can also identify problems before they become serious. Most problems can be
minimized if they are detected and repaired early. Normal mechanical failure modes degrade at
a speed directly proportional to their severity. If the problem is detected early, major repairs
can be prevented, in most instances.
Fig 2 . Different type of Maintenance
Implementing a “Best Practices” Predictive
Maintenance Program:
Avoiding the 10 Most Common Pitfalls
In an effort to increase equipment reliability and reduce unscheduled downtime, many
organizations have taken the proactive step of implementing a Predictive Maintenance (PdM)
Program. Unfortunately, only an estimated 20% of these initiatives actually achieve the
anticipated results, but knowing how to avoid the ten most common pitfalls substantially
improves PdM results.

In a global marketplace, it is imperative that industrial and manufacturing organizations operate

as competitively as possible. Maintenance has an enormous impact on an organization. It
influences equipment reliability, equipment availability, production throughput, and eventual
bottom-line profitability, so it is critical that informative and cost-effective equipment
management strategies be employed. PdM/Condition Monitoring, if implemented properly, is a
very effective strategy.

1. Improper Equipment Selection

To identify equipment health and then increase plant equipment reliability, it is important to
select the right assets to monitor. One major mistake that companies can make is skipping an
Equipment Criticality Analysis (ECA) and a Reliability Centered Maintenance (RCM) failure
management strategy. Bypassing these two key steps can result in overlooking plant equipment
that should be tested on a regular basis and failing to apply additional Condition Monitoring
technologies to assets where one technology may not be appropriate for mitigating all failure
modes. To ensure the team has a comprehensive list of machines that must be monitored with
various predictive technologies, it is critical to have a populated CMMS/EAM database with an
associated Equipment Criticality Ranking (ECR) or, at minimum, a simple list of the equipment
register sorted in a ranked order. The best approach is to go through
a formalized RCM analysis and identify opportunities to mitigate failures by using PdM
technologies on various plant equipment.

This effort will capture the assets that are candidates for the various technologies and reduce
unnecessary PMs associated with those assets.
 2. Improper Personnel Selection
A common mistake is selecting the wrong personnel to conduct PdM testing and analysis. When
staffing for an internal program, many organizations hastily assemble a team of their best
firefighting mechanics—individuals who have “saved the day” over the years. Although their
loyalty is typically very commendable, there are many other key attributes to consider when
selecting staff for an internal PdM program. Some of those characteristics include:

A diverse familiarity with plant equipment and system functions

„„Some degree of mechanical or electrical aptitude based on the PdM technology being utilized

„„Detail oriented Analytical in nature

„„Excellent problem solver

„„Computer savvy

„„Improvement oriented

„„Documentation and metrics driven

„„Seeks training and self development

3. Providing Inadequate Training

Often new PdM technicians receive little or no training beyond traditional vendor system
“button-pushing” training. In fact, many PdM specialists are starved for valuable training that
directly impacts their effectiveness and the success of the PdM program. Ideally PdM
technicians should receive the following initial training:

„„Basic introduction to predictive technologies

„„Equipment reliability strategies in general

„„Test instrument hardware training

„„PdM software training

„„Report writing training

„„CMMS/EAM user training

Future training should encompass:

„„Advanced PdM application technologies

„„RCFA training

„„Certification training after ample time has been spent in the field collecting data and
performing analysis

„„Power transmission/rolling element bearing

„„Precision shaft alignment

„„Precision balancing

„„Lubrication practices

For the Condition Monitoring program to succeed, it is critical that the personnel performing
the PdM data collection and analysis are properly trained. A shortfall in this area will directly
impact the quality of the overall initiative.

4. Lack of Repeatability
Repeatability leads to reliability, so it is important to ensure your program is based on
consistency. There are many potential variables that can impact the accuracy of the data being
collected, thereby affecting the accuracy of the analysis/recommendations on machinery health
and related corrective actions that may be taken. Some of the variables that should remain
consistent are:

Collecting data in the same location (utilize a paint pen or pad to measure at consistent
points)

„„Collecting data in the same surface conditions (clean surface, free from dirt and debris)

„„Utilizing the same instrument settings when testing the same equipment

„„Ensuring machinery being tested is loaded at the same operating parameters from test to
test

5. Not Collecting and Analyzing Data in a Timely Manner

Although this item may seem obvious, many programs do not keep up with the site’s data
collection activities or are delinquent in performing the associated analysis. An inconsistent
effort in collecting and analyzing data will lead to unidentified equipment failures and
unscheduled downtime, resulting in negative program publicity. A few of the more common
reasons for this occurring include:

„„Absence of establishing PdM routes

„„Lack of established PdM work orders coming out of the CMMS/EAM system

„„Poor planning/scheduling practices

„„Little-to-no review of delinquent PdM work orders

„„Lack in proper training of the PdM technician/analyst causing them to be hesitant in

performing their tasks

„„Lack of motivation or understanding of importance by the data collector/analyst of their

tasks
„„Lack of accountability in performing PdM tasks in a timely manner

„„PdM crew repetitively being pulled off their Condition Monitoring duties to aid in firefighting
plant “emergencies”

It is important to be protective of your Condition Monitoring personnel and to ensure that they
are performing PdM tasks that will contribute to the success of the program. Be aware of the
many gremlins that will inhibit them from fulfilling their responsibilities.
Do not allow your data collectors to become dust collectors!

6. Not Taking Corrective Action

If there is anything worse than not collecting and analyzing the data in a timely manner, it is
receiving the support and funding to develop a PdM program and identifying the proper
equipment to monitor with the ideal PdM technology, used by the best trained technician in a
timely manner and accurately analyzed by a qualified specialist and then simply taking no
action on identified equipment degradation conditions. Your program will be sure to come
under justifiable scrutiny as machinery anomalies are flagged and no corrective action occurs,
allowing critical plant assets to fail in spite of having the information required to proactively
take mitigating action. Ideally this would never take place, but here are a few reasons why this
may occur:
„„A process for converting identified equipment to be reconditioned or repaired has not been
established (The work request or work order process for PdM work has not been developed or
is not clear to those involved in the process)
„„
„„
7. Failing to Measure and Document Program Successes

A frequent deficiency in PdM programs is the lack of predictive metrics that are captured and
reported. This includes shortfalls in documenting and communicating the overall program wins
or successes. It is common to fail to identify
predictive KPIs, so it is important to perform the following exercise when measuring program
performance:
„„
Identify the PdM program metrics to be measured

„„Identify the target goals for each metric

„„Identify the individual(s) who will measure and report each key performance indicator

„„Agree on the formulas to use in how the metrics will be measured

„„Select the frequency of measurement

„Agree on the target audience to report this information to

„„Select the ways in which to communicate

There is often inconsistent documentation of program successes. Although some organizations

initially report program wins, many companies quickly falter and no longer document the good
things that the program is providing to the organization. For this reason, it is strongly
recommended that a consistent template be created to document program successes. This
PdM Program Scorecard should capture the number and nature of events that were realized as
a result of having an effective Condition Monitoring program in place as well as the annualized
cost savings that are achieved. It is important to quantify the amount of savings realized, so it
may be necessary to sit down with some of your financial and senior leadership staff members
to accurately identify what dollar savings
should be reported. Regardless of the template format, it is important to consistently provide
this document to those who are in a position of influence over the future of your program.

8. Lack of Program Support

Perhaps one of the most recognizable elements required to develop a successful PdM initiative
is program support. Having a program sponsor at the beginning of the process and throughout
the life of the program is essential. The following items represent some of the areas in which a
program sponsor adds value:

„„Providing a vision of what the PdM program could be at your facility

„„Assisting in selling the benefits of the predictive initiative to senior leadership

„„Constructing the budget and acquiring the funds to take the program from conception

„„If an internally staffed program, ensuring the program receives the personnel, test
equipment, training, tools (work stations, computers, etc.), and time to execute the program
successfully If pursuing an external program model, ensuring the program spells out the service
provider requirements as well as seeing that proper contractor management is being
performed.

„9. Only Using PdM as an Equipment Health Tool

One of the downfalls that often occurs is solely using predictive testing as a measure of
machine condition. The capabilities of PdM technologies are extremely diverse, and Condition
Monitoring teams are generally required to prove the value of what their group brings to the
organization, so it is important to utilize these tools in a number of different areas. Some of the
more common applications include:

„„Machinery health/Condition Monitoring of installed equipment

„„Performing incoming, acceptance testing of repaired equipment

„„Conducting incoming, acceptance testing of new equipment and lubricants

„„Performing motor circuit analysis on critical spare motors in the storeroom/warehouse

10. Not Adopting a Continuous Improvement Mindset

The last common pitfall is a stagnant Condition Monitoring program. One of the keys to a
successful program is that it must be dynamic in nature. As things change over time, so does
the need to update your PdM program elements. Key components of a continuous
improvement effort include program leadership, periodic reviews, performance metrics, and
data to analyze change over time. A few of the items that commonly need to be reviewed are
:
„„The equipment to be monitored: As old equipment is retired and new equipment is
commissioned, your list of equipment to test will change. This is also true for assets that were
not originally on the list until an equipment failure proved that an asset needs ongoing
monitoring.
„„The frequency of monitoring: As some assets enter into alarm conditions or experience latter
stages of failure, it may become necessary to increase the monitoring frequencies. On the other
hand, if several months of monitoring less critical machinery shows no indication of machine
wear whatsoever, the decision may be made to moderately extend the testing frequency
intervals.
„„The test instruments being used: As hardware and software programs change and improve, a
different brand of instrument or simply a newer instrument with much-needed enhanced
capabilities may be selected.

Fig 3 . complexity and potential of PDM„„

Case study in chemical Industries :
There is an impressive case study from JMP from chemical development: A company wanted to
bring a catalyst to market that could synthesize aliphatic polycarbonate polyols from waste
carbon dioxide. 35 factors were identified that might affect yield, polymerization rate and by-
product production and they couldn’t see an efficient way forward to optimize the process.

In the past they had tried traditional experimental – or one factor at a time, OFAT – approaches
with varying success. Sometimes they were lucky and identified a solution quickly, other times
they were unlucky and had significant over-runs. They heaved a sigh of relief that a solution was
found, but were never sure they had found the best, most robust, or lowest cost process to
operate.

By using JMP Pro they were able to exploit their existing data using data mining to identify the
top 10 factors out of the 35 that might be responsible for their KPI outcomes. Using these top
factors, they defined an efficient data collection plan requiring 21 individual experiments.

After conducting these experiments, they were able to fit a predictive model, query and present
that model graphically to gain insight and communicate that insight to key stakeholders,
delivering the solution needed to successfully scale production to 7,500 liter capacity. The
findings were at odds with simple kinetic theory which also led to a better understanding of the
reaction mechanism. Rather than commercializing the catalyst, the business unit was spun off
to Saudi Aramco for $100 million.
Advantages of Digital Transformation :
On the whole the chemical industry has been good at the adoption of improvement techniques
– from understanding plant and performance and adopting measures like Overall Equipment
Effectiveness to the use of the techniques of Deeming such as Kaizen and Six Sigma. However
even within these methodologies there is still a tendency to allow anecdote and the pressure of
personality to be perpetuated. Once the low hanging fruits have been taken using these
techniques, they sometimes become a means to manage and can lose momentum.

All those that use these techniques need greater statistical insight into the data that they
generate. Data analytics can provide more accurate insight, really showing what is important to
control. Sometimes more importantly they can reveal the many things that do not contribute to
better outcomes, thereby allowing people to focus on the things that really matter.
Furthermore, the modelling within the software will enable them to test their ideas before
making real process changes. This has not been possible outside of the factory, laboratory or
workshop until now.

The key to the use of statistical software is to have data normalized in some way. A date or time
marker, batch number even in continuous processes data can usually be associated with a
periodicity. With a limited amount of effort, managers, scientists and engineers will be able to
identify relationships and show real statistical evidence of those connections. With that
knowledge they will be more encouraged and motivated to take a look into the depths of their
data historians and make the effort to analyze the data within. Data can be input from many
formats. There can be no doubt that by improving the understanding of the valuable insights
that can be gained by using analytical tools, this should be enough of a driver for most to put in
the effort to normalize their data. Especially when it will enable improved testing, monitoring
and delivery of the performance of their processes, plant and equipment and giving them
access to their hidden factory.

New chemicals, intermediates, processes and formulations are being developed in these
industries on a daily basis and getting them to market more quickly means more value could be
realized earlier. The key challenges are always to minimize laboratory and pilot plant time
whilst gaining the most information from the work undertaken. Establishing a new product or
capturing sales because you have responded best to an enquiry or an opportunity is a crucial
commercial driver. This is where analytics can have such a valuable impact on innovation
workstreams.

In addition to quickly observing key process relationships that are statistically significant,
advances in analytics software to aid in the “design of experiments” can help speed up decision
making in chemical development. Such smart experimental design will quickly identify the
parameters that need to be tested to prove their potential sensitivity. This ability to focus on
what matters is key to limiting the amount of experimentation that has to be done to capture
enough data to be meaningful. This level of understanding enables innovation leaders and R&D
managers to be more precise about the time they will need to deliver their results.

"In the environment of Industry 4.0, maintenance should do

much more than merely preventing downtimes of individual
assets. Predicting failures via advanced analytics can
increase equipment uptime by up to 20%."
There are many sources of data in manufacturing operations and just as much if not more in
maintenance activities. Often, process equipment-based data is ignored because there can be
too much of it, and spotting important relationships is difficult. These relationships can arise in
the data from the equipment itself but also from the surrounding processes and operational
data. Data is generated by sensors, measurement systems, process control systems and process
inputs and outputs; complexity is compounded by such data being oftentimes collected with
different periodicities.

“Too many factories are producing

a mountain of data that is
never used.”

Too many factories are producing a mountain of data that is never used. Companies are starting
to realize – especially with initiatives like Industry 4.0 – that they’ve paid for all this data and
they need more returns from it. It is an asset in which many hidden gems may lay. There may
be real value hidden about how to improve the process, or how to make the product better or
more efficient, but they need the tools to release it.

According to McKinsey research, implementing

predictive maintenance using IoT can reduce
equipment downtime by as much as 50 percent
and reduce maintenance costs of factory
equipment by 10 to 40 percent.
Short cums (Challenges) :
In the chemicals industry, like many others, there is considerable excitement about the
potential of advanced predictive-maintenance (PdM) approaches. The promise of these new
techniques is tantalizing. Using machine-learning technologies to comb through historical
performance and failure data, they aim to tell operators when and how a component is likely to
go wrong in the future with a high level of predictability. This should reduce the impact of
equipment failures—and the cost of efforts to prevent such failures—by turning inefficient,
unplanned maintenance activities into efficient, planned ones.

At first sight, chemicals plants seem like the ideal environment for PdM. High levels of
automation and instrumentation, combined with rigorous maintenance record-keeping, create
the rich data that machine-learning systems require. Moreover, most plants strive for stable
operating conditions, potentially making it easier to spot patterns and trends. There’s also a
compelling business case for improved reliability. Overall equipment effectiveness (OEE) losses
due to unplanned maintenance range from 3 to 5 percent across the industry.

Predicting poor results

Take a closer look, however, and the potential of PdM in chemicals begins to evaporate, for four main
reasons.

— Too little data

In a chemicals plant, predicting

failures is harder than it first appears. Unplanned
downtime is typically concentrated in a small
number of large events. That means there are
typically too few datapoints for PdM systems to
learn from.
— Too little time.

Even when it’s possible to create

models with predictive power, they often work
over time horizons that are too short to be useful
in chemicals manufacturing. Predicting that a
part will fail in two days or two weeks is useful
in a truck or machine tool, but it may not help
in a plant where shutdowns take several days
and maintenance teams require months to plan
interventions and source spare parts.

— Too little impact.

The impact from PdM is often

low because plants operate critical assets with
a high degree of redundancy and few single
points of failure. If a pump stops unexpectedly,
operators can often switch to a backup unit
with little impact on production.

— Too little savings.

Finally, a focus on reducing

unplanned downtime ignores the largest
source of throughput losses in most plants.
Shutdowns for planned maintenance events
cause OEE losses of 5 to 10 percent on average,
twice as much as unplanned stoppages.

“A long-term strategy for adequate model updates on

failure prediction is crucial.”
Towards digital reliability
Do these challenges mean analytics provides little or no value in efforts to improve asset
productivity in the chemicals sector? No. The industry is achieving considerable success with a
range of digital reliability techniques, many of which are far cheaper and less complex to
implement than advanced PdM.

Condition monitoring

Improving condition monitoring through better remote sensing can cut mean time-to-repair,
significantly reducing the impact of equipment failures. At one chemical plant, a few critical
pumps suffered repeated failures. No backups were available for these units, and the issue was
a significant source of production losses at the plant. “We decided we couldn’t wait for the
plant and reliability engineers to identify the root cause, redefine the pump’s technical
specifications, and then procure replacements,” the plant’s maintenance manager told us. “So,
we focused on mitigating the impact of the failures, rather than avoiding them.”
The plant’s reliability team installed a handful of new sensors on the pumps and started to
monitor their condition online in real time, allowing them to detect imminent failures a few
hours before they occurred .By enabling maintenance personnel to be ready to intervene, this
intervention reduced the mean time-to-repair on these pumps from 6.5to around 3 hours,
cutting OEE losses by almost half and saving approximately $120,000 for each failure.

Root-cause problem solving

Better data also means better root-cause problem solving. That helps companies to prevent the
recurrence of failures, to improve their failure-modes and- effects analysis (FMEA) processes,
and to optimize preventative-maintenance plans. Together, those actions address the critical
aspects of reliability performance, reducing both the impact of failures and the cost of
preventing them.
At one chemicals plant, for example, failures in a critical piece of equipment caused operators
to activate an emergency shutdown three times in as many months. These shutdowns were
inconvenient enough—but when the site team attempted to restart the plant, they found that
the abrupt shutdown of the unit led to an accumulation of solids in key vessels and pipes. Fixing
that problem led to lengthy delays in start-up and significant losses in output.
To address the issue, the company applied a combination of traditional root-cause problem
solving and smart analytical techniques. Analysis of process data helped them understand how
and why solids were accumulating under emergency shutdown conditions. The issue was fixed
with a combination of enhanced monitoring and changes to preventative maintenance plans.
But the data driven insights also allowed the plant to revise its emergency shutdown
procedures to stop the plant safely without causing the solids problem. That
change reduced start-up time after any kind of emergency shutdown by 90 percent.

Conclusion
The potential for digital reliability extends far beyond predictive maintenance. And for
chemicals companies, we believe that these other digital approaches are both easier to
implement and offer greater value. The highly instrumented nature of most chemical
production facilities means many companies already have a rich, and largely untapped, source
of data to support digital reliability efforts. For those plants that still don’t, then it’s time to
“sensor up”: better data is the vital first step on the digital-reliability journey.

IoT and the power of advanced analytics and machine learning is shifting the paradigm in
predictive maintenance today. Utilizing the power of Data Hub, organizations can now easily
ingest, store, process, and analyze unlimited volumes and varieties of sensor data; use powerful
processing and analytics tools across data in motion as well as data at rest; and leverage
machine learning capabilities across petabytes of data to drive real-time predictive
maintenance. Organizations are able to benefit from the power of open source technologies
while leveraging industry-leading management and data security tools that are critical to IoT
production deployments. Today IIOT is powering some of the most compelling IoT use cases
across diverse verticals, and, together with our industry-leading partners, can accelerate the
time-to-value for your IoT investments.

"Predictive maintenance could help you manage

maintenance more efficiently. However, keep in
mind that not all enterprises require the same level
of reliability from their assets."
Reference

1 Predictive Analytics with MATLAB

2 2 DRIVING PREDICTIVE MAINTENANCE IN A CONNECTED WORLD
By Clodera

3 Predictive Maintenance –Delolite

4 Implementing a “Best Practices” Predictive Maintenance Program

- emerson
5 Predictive maintenance :the wrong solution to the right
problem.- Mcknesy & Company