SPE-194962-MS

Drilling Process Digitalization Using Advanced Machine Learning

Techniques – Case Study

Adel Al Shayaa and Khaled Al Tamimi, Mubadala Petroleum; Sara Bakhti and Dr. Arghad Arnaout, TDE
International Limited; Dr. Gerhard Thonhauser, Montanuniversitaet Leoben

Copyright 2019, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Middle East Oil and Gas Show and Conference held in Manama, Bahrain, 18-21 March 2019.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
The digital era is upon us in the oil and gas industry, enforcing the importance of integrating this novel
concept of digital transformation is mandatory to improve the overall efficiency of the drilling process.
The advances in Artificial Intelligence for the industry, and the methodology of how these solutions can be
implemented will be addressed in this paper.
The digital transformation of drilling will provide an unprecedented stream of high-quality information
that has never been accomplished in the industry, through the utilization of automated real-time drilling
downhole tools, data analytics and predictive analysis.
Therefore, a real-time measurement and processing technology based on automated rig activities
detection was established to improve the performance of the rig crew and drilling operations of an
international operating company. A monitoring process was implemented for all drilling operations and
benchmarked against operational Key Performance Indicators (KPIs).

Introduction
Currently the oil and gas industry seeking to increase the involvement of new advances in digitalization
technologies at different sectors. The big challenge is how the digitalization of the workflow of drilling
process can be done. When performed, it opens the doors wide to increase the engagement of different firm
levels in decision making process and make it more reliable and justifiable.
The first step towards improving digitalizing is to establish a system that collects and aggregate the data
from various sources. Usually rig sensor data is aggregated and stored at the rig site and then transferred to
the office. If the data of a poor quality, then it makes no sense to start with any monitoring or improvement
process.
Then, A digitizing step is essential at this phase to convert received raw data with high quality into a
digital status based on which we can build further decision models using computing powers.
At that point, the integration step of this information from multiple sources to one defined point - the
operating headquarters, allows the corporation to track the performance and establish goals.
2 SPE-194962-MS

Figure 1—Example of Digitalization Process Workflow.

This digitalization process is an enabler to increase fleet efficiency and drill more wells in less time
through monitoring and control operations for the different business units. It can be achieved by setting
number of KPIs that are being used to assess the drilling performance with very short time intervals. This
gives a clear idea on the difference between drilling crews and different rigs in executing similar drilling
operations using same equipment. Measuring drilling performance from operational sensor data is the first
basic step towards improvement.
An essential step implementing a systematic approach of performance improvement and communicate
the performance gap to the drilling crews. This highly affects the performance improvement process and
maximize the outcomes of measuring the well delivery process.

Continuous Improvement Process

The process shown in Figure 2 is adopted by the company to assess the current operational performance level
and translate all captured lessons into real actions. The systematic approach for improvement is required to
make the result of the improvement process sustainable and increase the expected value.
SPE-194962-MS 3

Figure 2—Improvement process

The process consists of four main steps: Check, Act, Plan and Do.
1. Check is the first step that is required to measure and evaluate the current situation of the drilling
process. Automated reports generation and calculating the Key Performance Indicators (KPIs) are
the main tasks to give a clear evaluation and assessment of the drilling process.
2. Act is important to evaluate the results of the measurement step and to determine positives and
negatives in the current workflows, to decide what is required to improve the situation.
3. Plan is an essential stage to prepare for the required implementations. The required actions
concluded from the previous steps are converted into input for well planning and scheduling where
the KPIs Targets, Lessons Learned and Best Practices are now part of the plan.
4. Do is the execution phase of the process. In this step, all the planned actions should be executed and
reflected on the current workflows and coaching the crews to know how to react and run the plan.

Data Quality Control

Accurate data and efficient quality control is the backbone for performance improvement process. High
data quality is the base for a fair evaluation of performance and to properly select valuable and trustable
improvement initiatives. Usually rig sensor data is aggregated and stored at the rig site and then transferred
to the office. If the quality of the data is not up to the level to make the performance measurement possible,
then it makes no sense to start with any monitoring or improvement process.
A special set of quality KPIs is used to carefully evaluate and measure the quality of aggregated data.
These KPIs are designed to directly reflect the quality of data and provide a direct feedback to different
parties in case data with poor quality was received. The data quality control KPIs are:
1. Validity: represents the ratio of valid received data values, the valid value is within a pre-defined
domain specified by engineers. For example, block position values can be between 0 and 90 meters
based on the rig derrick, any value located outside this range can be considered as invalid.
2. Accuracy: shows the ratio of accurate received data, the accurate data reflects the exactness of
measurements performed by the measurement system.
3. Consistency: measures the degree of uniformity of received data, the consistency of the received data
gives an idea on the regularity of the data received, for example, the ratio of duplicated records in
the received data.
4 SPE-194962-MS

4. Integrity: shows the ratio of valid consistent relations between different sensors, for example, if the
drill string is "In Slips" state and bit depth changes during this state, other examples are RPM with no
torque readings and pump strokes without any mud flow-in rate.
5. Timeliness: indicates whether the data is available at time needed.
6. Completeness: checks whether all necessary data received as it was expected.
7. Continuity: represents the regularity of received data or regularity of gaps in the data, are the gaps
received regularly or randomly.

Process Workflow Implementation

The technology of real-time performance measurement tools is based on automated rig activities detection
and classification from rig sensor data. This technology is the fundamental base for evaluating the
performance. It uses the different high-tech and advanced machine learning techniques to automatically
recognize the different data patterns that are generated by rig equipment while the crews drill a well.
The next step after finalizing the data quality insurance assurance is to setup the workflow and how the
measurement process will power the current existing workflows with the right information at the right time.
There is nothing more powerful than sharing high-quality information between all involved parties for the
drilling process. Through effective communication, everybody knows the reality and how to react based on
the current situation and team decisions.
The company shall apply a workflow for sharing information between operators and drilling crews to
create a common base to understand the operational performance level. Such information should be fair and
independent. The unbiased high-quality information is a key success to determine performance shortages
and improve the work efficiency. In some perfect scenarios, the organization that provides the performance
measurements services should be independent from the any other service provider. Establishing a drilling
performance measurement process supports the company to evaluate the contractor performance and give
the suitable feedback at the right time.
As a next phase, an advanced performance measurement tool is hired by the company to translate
the sensor data and convert it into objective information of the drilling performance. As initial step and
before starting with the implementation of an information-sharing process, the teams create shared reporting
templates that can be used to improve the current workflow and power it with the right performance
information. A series of meetings shall be held to raise the awareness level for all the participants.
Figure 3 shows an example of one of the daily reports that were created and generated using such
technology on daily basis. These reports are sent to the drilling crews, as well as the company drilling team
at the same time.
SPE-194962-MS 5

Figure 3—An example of automated daily report

Automated Drilling Performance Measurement

One of the next major tasks to be solved is how performance information of each drilling crew for different
operations can be generated out of the automated daily reports.
The main condition to make these KPIs meaningful to drilling crews is to report them exactly at the
time when they take place. This process of information fusing has a great value to drilling crews in terms
of measuring the performance and communicating each abnormal performance with the others, to find a
common solution and to exchange the expertise.
Different Key Performance Indicators are used to monitor the drilling process during different jobs such
as tripping, drilling formations, running casing, etc.
Figure 3 shows an example of how the performance of each crew is reported using a "Slip to Slip KPI".
A "Slip to Slip" operation is the time from putting a string in slips to the time when the drilling string is
out of slips.
6 SPE-194962-MS

Figure 4—chart shows Slip-to-Slip KPI histogram.

Figure 3 Slip-to-Slip KPI histogram. In this histogram, the operations count (number of needles) is visible
on the Y axis and time ranges in terms of operation durations (needle duration) on the X axis. So, the green
bar of the left histogram means that 10% of the operations for Slip to Slip connections were made in the time
range of 1.5 to 1.7 minutes. The red line represents the target that has been set for this Key Performance
Indicator.
Three steps of improvement:
1. The distribution of timings for specific events have been analysed and targets set using the existing
Best Practice on the rig, shown on the histogram on the left – target 2 min (close or on the 50%
mark).
2. Improvement of the operational consistency by narrowing the distribution by more consistent
operations around the Best Practice shown on the histogram on the right – tighter bell curve.
3. The operational performance improves by learning and adopting Best Practices, so that the target
shifts to the left gradually over time along with the bars expressed with the arrows in the histogram
on the right – move from the orange towards the blue curve.
All operations that took longer than the target (bars right of it) are Invisible Lost Time. The shape of
the histogram can be used to assess consistency and the potential of target development over time. The
implementation of a performance culture and setting motivating competitive targets among crews complete
the efforts for performance improvement. This analysis can be done for all KPIs to identify the drivers of
savings potential.
Maybe there is a good reason to explain why the connections took longer than expected. But just sharing
this information between the parties helps the company to put the performance under control and support
the teams that sometimes suffer from a shortage in resources or face some technical difficulties to reach the
best possible and safest performance.
Using this technology for reporting drilling performance supports the company to develop its drilling
contracts and make the relation with drilling contractors based on and measured by performance KPIs. By
setting targets using different KPIs, the company can measure the performance and can identify all the
SPE-194962-MS 7

deviations from the expected performance (the benchmark). This also supports the contractor if a crew needs
any support or when facing some technical difficulties.

Invisible Lost Time Mitigation

This inconsistency of performing the same operations by crews using same equipment, is the main source
of the Invisible Lost Time (ILT). The ILT can be defined as the difference between the actual operation
duration and the benchmark. Mitigation of ILT can be done through the following:
▪ Continuous measurement of the drilling performance.
▪ Communicate the information on the performance to drilling crews.
▪ Develop recommendation on best practice for different operations by focusing on small set of KPIs
▪ Rig-site support and coaching by support engineer on the rig-site to establish process
Figure 5 shows an example of a performance gap analysis indicating the measured ILT. After the
inspection, it has been found that the main contributor to the inconsistency in performing the operation is
the Tripping "Slip to Slip" operation. A similar, partly inconsistent performance trend for day and night
crews was spotted for the analysed Drilling weight to weight times.

Figure 5—Performance Gap Analysis

ILT during drilling could be reduced by more consistent weight to weight connection times as well as by
proper wellbore treatment & conditioning timing. The hole cleaning procedure was not standardized, and
each crew performed it on its own. Training the Crews on working with the support of automated drilling
report would help to increase consistency while performing Slip to Slip Connections both during Tripping
and Running Casing.

Digitalization Transformation
Using historical drilling data, will enable AI technology to calculate future predictions, to learn from the
past to improve the future. The digitalisation of drilling process has the ability to sustain a constant flow of
information throughout the internal structure of this operating company and its unique setup.
Follow up the defined goals and track the improvement and efficiency over time and find a performance
gap to establish a feedback link to correct the deviations.
8 SPE-194962-MS

The eventual goal is establishing and improving the global process and workflow of monitoring the
operating units for the multi-level operating team structure, by the means of high speed, digital data
technologies over considerable slower human factors.
The performance can be improved dramatically after Setting targets on a KPI basis. These KPIs targets
should be communicated to the crew and tracked on a daily. This would not only increase consistency but
also it could significantly improve safety.
An important stage in the performance improvement process is to discuss on a daily basis the rig crew
feedback & lessons learned, compare the planned and actual operations as part of the morning meetings and
update the well operations prognosis on a daily basis. The interaction between company team and drilling
crews based on accurate and unbiased information that is provided at the right time is the main contributor
for performance improvement.
As shown in Figure 6, Management can easily track the rigs’ performance through a management
dashboard that shows performance trends for the entire rig fleet monitored across different business units.

Figure 6—Management scoreboard for efficiency monitoring

Acknowledgements
The authors would like to acknowledge proNova TDE team for their efforts to provide performance reports
in the required formats at the right time.