CONDITION MONITORING LOCATION (CML) OPTIMAIZATION:

PART-1: INTRODUCTION.

Since when the oil & gas and petrochemical industry enforced to monitor and capture
the metal losses and recoding the wall thickness over the time to keep the plant and
sites safe. NDT techniques also evolved with time, however the process of sending
NDT technicians into plants to collect and manage the data has not changed as much
over the decades. It is a laborious work and depends quite number of factors such as
highly variable work environments, safety risks and data ambiguity issues.

Furthermore; managing all the data, both good and ambiguous, continues to be very
challenging. Accurate manual thickness readings and trends are also challenging to
monitor due to variability in technician qualification and experience, temperature
changes, instrument differences, surface conditions, calibration and measurements
not taken on the same exact spot. Sometimes it has been observed that there is wall
thickness gain between two readings. This shows that a more robust solution is
required, especially for trending highly critical CMLs/TMLs (thickness monitoring
locations). Even if accurate readings are taken at some intervals, some events
caused by process upsets or turnaround related issues will not be understood
properly. The role of corrosion engineer is really very important in mapping such
metal losses against the thickness taken, they can identify such uncertainties in the
measurement readings and correlate these with actual process parameters and
IOWs.
PART-2: CONCEPT OF OPTIMIZATION:

To grasp the concept of optimization, it is really needs to understand the uncertainty

of the models used and their relationship with scaling the accuracy. It can be started
with more qualitative or semi-quantitative modeling where there is lack of
information or data (high uncertain bands), producing higher risks or more
conservative results. On the other hand, fully quantitative modeling where there is
more data is available, more statistical calculation models are available, which
usually produce more accurate results with less scatter (smaller uncertain bands) in
the prediction.

The graphic below shows how data can drive the model to be conservative and/or
non-conservative. If the data is ambiguous and scattered, we will be having more
conservative model, in this case, although it is safe, however producing high
uncertain risks and un-wanted results. Whereas non-Conservative model is not
desirable since it is not safe. As equipment grows older or more damaged, we will get
more and data to tell us more about the condition of the equipment and we will come
closer and closer to the absolute certainty line about the true damage state or
condition of the equipment.

Data science can help in predicting the results based on available data and past
inspection records. This could reduce the band of uncertainty significantly if the
model adopted trained based on the actual conditions and with the help of right asset
integrity professionals. It could also help for aging assets to accurately conduct FFS
assessments.

PART-3: PIPING CIRCUITIZATION REVIEW:

Subsection of piping systems that includes piping and components which are
exposed to a process environment of similar corrosivity and expected damage
mechanisms and is of similar design conditions and construction material, whereby
the expected type and rate of damage can reasonably be expected to be the same.
Complex process units or piping systems are divided into piping circuits to manage
the necessary inspections, calculations, and recordkeeping. When establishing the
boundary of a particular piping circuit, the inspector may also size it to provide a
practical package for recordkeeping and performing field inspection. By identifying
environments and damage mechanisms as circuits, the spread of calculated corrosion
rates of the CMLs in each circuit is reduced. Proper selection of components in the
piping circuit and the number of CMLs are particularly important when using
statistical methods to assess corrosion rates and remaining life.

To establish and review piping circuitization, start with a very basic step by
considering and identifying Piping Systemization. API 970 is a good guide for
establishing corrosion systems. API 571 will be helpful for identification of respective
potential damage mechanisms. All process piping systems shall be categorized into
different piping classes except for piping that has been planned based on RBI. Such a
classification system allows extra inspection efforts to be focused on piping systems
that may have the highest potential consequences if failure or loss of containment
should occur. API 570 and API 574 will be good guidelines for establishing the circuits.
Inspection history review will be very important which can certainly identify probable
location of damage.

PART-4: PIPING CIRCUITIZATION OPTIMIZATION:

Piping circuits can be optimized based on the risk projection and risk ranking. Risk-
based inspection (RBI) assessment is a good methodology and tool to identify the risk
across a facility. Industry standards like API 570 and API 574 can help in placing them
at the right locations. CMLs optimization can further help in reduction and
optimization of piping circuits and guide where to focus more.

Optimization of CMLs can be accomplished either based on the risk projection or

application of AI models such as data science, data analytics, machine learning, and
deep learning etc.
PART-5: PERFORM DAMAGE MECHANISM REVIEW (DMR):

A damage mechanism review (DMR) is performed to establish corrosion rates,

credible damage mechanisms, and susceptibility to damage for each component,
group of components (e.g. circuit) or equipment items. The damage mode (e.g.
thinning, pitting, cracking) and failure mode (e.g. pit-to-leak, crack-to- leak) may also
be identified during the DMR. The information derived from the DMR is essential to
the quality and the effectiveness of the risk assessment since it is used to estimate
the POF. The DMR mainly addresses the factors that affect the POF; but the likely
mode of failure indicated in the DMR may also affect the COF.

The DMR process includes a systematic review of the design data, process
information, and materials of construction along with available inspection,
monitoring, and sampling data. The corrosion specialist shall define the credible
damage mechanisms for each item addressed by the RBI assessment. Damage
modes (which may be optional in some RBI approaches) should also be defined for
those same items. Process specialists and operations personnel can provide useful
input (such as the spectrum of process conditions, injection points, etc.) to aid
corrosion specialists in the identification of credible damage mechanisms and rates.
For example, understanding that localized thinning may be caused by the method of
fluid injection and agitation may be as important as knowing the corrosion
mechanism.
PART-6: INSPECTION DATA MANAGEMENT SYSTEM (IDMS) REVIEW AND
VALIDATION:
Inspection Data Management System (IDMS) is a software program or system
designed to collect, store, organize, and analyze data generated from equipment
inspections and testing in industrial facilities. The dataset within an IDMS includes
crucial information about asset conditions, performance records, maintenance
history, as well as detailed inspection reports, test results, and interconnected data.
The primary purpose of an IDMS is to streamline and consolidate all relevant
information related to a facility's equipment inspection processes. By incorporating
IDMS systems into their operations, organizations can significantly enhance the
efficiency and effectiveness of their asset management processes.

IDMS data is “gold” for the asset integrity and risk assessment experts, therefore
review of and validation of thickness measurement readings with existing CMLs
should be conducted. It is really very important for the the people conducting RBI
assessments and related studies to ensure that CMLs Groups are created accurately
and respective CMLs are allocated and marked accurately on inspection drawings,
where needed, a field or site verification should be carried out.