IADC/SPE-180651-MS

Integrating Software for Improved Well Planning

M. C. Ponsonby, AGR Australia Pty Ltd.; O. J. Andersen, and P. Mathisen, AGR

Copyright 2016, IADC/SPE Asia Pacific Drilling Technology Conference

This paper was prepared for presentation at the IADC/SPE Asia Pacific Drilling Technology Conference held in Singapore, 22–24 August 2016.

This paper was selected for presentation by an IADC/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 International Association of Drilling Contractors or the Society of Petroleum Engineers and are subject to
correction by the author(s). The material does not necessarily reflect any position of the International Association of Drilling Contractors or 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 International Association
of Drilling Contractors or 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 IADC/SPE copyright.

Abstract
Each company has its own method of capturing data during the Drilling and Well Construction process,
and these methods will vary from simple spreadsheets to more complex systems that captures the live data
and stores it. Each company will also have a method of capturing the lessons learned, usually in the form
or an End of Well Report or wash up meeting with all of the service companies. Turning this data into
useful information and thence to knowledge is at the moment a difficult problem
This transition process had been brought about by combining two specialist pieces of software. The
first has the ability to take in date from all sorts of sources ranging from spread sheets via daily reporting
systems to historical log data or real time data stream, mud logs etc. and then to present that data in a form
that is useful to the drilling engineer in designing a well. The software also allows storage of experiences,
both positive and negative, and links them to wells, rigs, specific types of equipment or operations.
This central store of data and the ability to transition it into useful information gives the engineer
designing a well the ability to identify the risks and also the spread of times taken for various operations.
The second piece of software uses probabilistic calculations and the ability to place branches into the
process that take into account the risks identified using the data storage software. Once a risk is identified
then the mitigation steps are identified: how it will be addressed, what is needed and when.
All this information is taken into account when calculating the time and cost, giving a spread from
Technical Limit through to Worst Case. More importantly all of the steps, and assumptions are stored in
one place rather than the traditional multiple spreadsheets that generate the ⬙Excel Hell⬙ scenario.
By following a simple process with only two pieces of software the accuracy of well costing and
scheduling can be increased additionally the understanding and mitigation of risk improves and lessons
learned are no longer lost as personnel move within the company or the industry.

Introduction
There are many different sources of data used when designing a well, quite often with duplication and
contradiction. A data analytic system was developed to compile all of the information in one system that
allows quality control and validation of the data. The quality control and validation has proved to be more
useful than originally conceived as the data is presented pictorially and missing items or incorrect items
are very quickly identified, such as missing casing sections or no cement data.
2 IADC/SPE-180651-MS

Figure 1—Area of Data Management

As the analytic system has developed its ability to import additional sources of information has
improved as well as the ability to work with the data and present it in a form that is usable for other
software packages. This standardisation ability allows engineers to access data in a common format and
present reports where the provenance may be easily reviewed.

Method

The normal method of planning a well starts with the Offset Well review where the drilling engineer
searches through different information sources for the important information. Figure 2 below shows some
of the sources of information that may be used for the report.

Figure 2—Offset Well Review Data Sources

The information regarding the casing depth data is usually presented using a stick diagram as shown
in Figure 3.
IADC/SPE-180651-MS 3

Figure 3—Traditional Stick Diagram

The additional data such as incidents that happened during the operation, lessons learned that are
applicable and other useful information are collated in a separate document or possibly in the same
spreadsheet that was used to record the information for the stick diagram.
Originally the system took in data from a spreadsheet template that covered:
● Bit Record
● Casing Scheme
● Lithostratigraphy
● Location Data
● Mud Details
● String Details
● Experiences (Lessons learned or incidents that took place)
The capability has been updated to import data from even more sources, including live drilling
information and existing databases such:
● IDS
● WellView
● OpenWells
● SiteCom
● Mud Logs and varicose types of Log ASCII Standard (LAS) files.
4 IADC/SPE-180651-MS

The ability to gather all of this data into one repository ensures that all the data used for the Offset Well
analysis and subsequent detailed engineering is kept in one place and can be referenced by engineers who
may take over the project.
The Offset Well review now takes on a set process that is repeatable and traceable for the source of the
information. The well selection process is quite simple and providing all the well data is in the system well
selection may be filtered by different standards such as Well Depth, Water Depth, or Rig type and the
selection is then saved ready for analysis.
Due to the large amount of information that the system is able to import and collate the well comparison
chart now becomes very detailed with far more relevant information that may be used. Figure 4 shows an
example high level of detail that is available for review in the diagram, it not only contains all the data
that would be available in a stick diagram but also includes information on the lithology, formation
strength tests tops of cement and more.

Figure 4 —Well Comparison Diagram

IADC/SPE-180651-MS 5

All of the wells selected are shown side by side with a common depth scale. The diagrams may be
sorted by pre-sets, such as rig, distance depth or dragged and dropped to a specific order. Each of the
diagrams can be exported as a PDF with all the well details included, or if required a batch export of up
to eight wells may be made.
The diagrams show details of the casing depth, formation tops, cement tops and FIT. If the mouse is
hovered over a casing string the information is summarised including information on the mud type.
Once the programme was in use it became apparent that there were sometimes additional sources of
information or documents that should be easily available for review. The programme now has the ability
to upload documents into each well file. This allows normal documents such as the End of Well Report
to be uploaded but also if there are specific reports generated following the operation these can be
uploaded and kept securely.
An important part of the Offset Review is reviewing any incidents and lessons learned and their
applicability to the operation that is being planned. Figure 4 shows three wells that are part of an Offset
Review, they have been grouped together as all have an Experience, denoted by a red thumbs down icon,
at the same depth. Hovering over the icon brings up the Experience information, in each case this
Experience has been a shallow gas issue. Clicking on the icon brings up all the details including a
Description, Immediate Solution and Future Recommendations.
Note the well on the right has green thumbs up icons, this denotes ideas that have been tried and shown
to be successful. In one of the cases it is success of drilling through a certain formation with a KCl-Glycol
mud system.
The experiences in the diagram all relate directly to the wellbore and the drilling operation, which only
covers part of the lessons learned. The Offset Review section of the programme also access information
on all experiences for the wells that have been selected, these include experiences related to the rig and
the equipment that was used, all of this may be exported into a separate document if required.
The risks identified in the review may now be used in the probabilistic planning software looking at the
cost and time of the operations. In normal deterministic planning each step of the project will be given
a set period of time that is used to calculate the cost, in probabilistic planning the step is given a range
of time that is as well as a type of distribution that the range will follow. Only three data points are
required for the information:
● Best Case
● Most Likely
● Worst Case
The most commonly used distributions are Pert, and Triangular.

Figure 5—Triangular distribution

6 IADC/SPE-180651-MS

The Triangular distribution approximates more complex distributions, such as the normal and lognor-
mal, by a simple triangle and is commonly used to model operations that are not undertaken on a regular
basis. The Mean and Standard Deviation of Triangular distributions are equally sensitive to the three
inputs.

Figure 6 —PERT Distribution

The Pert distribution is normally a skewed distribution that allows the duration of an activity to
occasionally assume vary large values but rarely assume very small values. It requires the same three input
parameters as Triangular. The Mean of the Pert distribution is four times more sensitive to Most Likely
than the Best or Worst Case.
The most powerful part of the probabilistic planning programme is its ability to take into account risks.
The risks that have been identified in the Offset Well review can be entered into the programme. These
can be simple risks as shown in Figure 7. This type of risk can be looked at as a diversion from the normal
path and is given a probability of occurring.

Figure 7—Simple Risk

IADC/SPE-180651-MS 7

More complex risk may be dealt with where there are multiple options as shown in Figure 8, each
option is given a probability of occurring.

Figure 8 —Complex Risk

A key benefit of P1 is ability to measure wells / sections / risks up against each other. The ideal is to
remove all risk, but quite often one has to go with the ⬙lesser of all evils⬙. Setting up several probabilistic
models with different designs clearly shows which of the models a lower risk profile with higher mean
and lower mean but more risk (typically)The ability to identify the risk in this fashion has several benefits
● they are all listed in the planning process
● the steps to mitigate the risk are addressed
● the impact on the duration of the operation and the associated costs are captured
Figure 9 shows how risk identification and quantification demonstrates the potential risk exposure for
an operation allowing an informed decision process to take place or for a re-focus on the engineering to
reduce the risk.

Figure 9 —Cost impact of Risk

8 IADC/SPE-180651-MS

The data analytic software is able to import very detailed drilling data from various industry standard
sources which allows the information from the offset wells to be used in the detailed engineering that
would go into the drilling programme. Normally choosing which bit to use by past performance would be
quite time consuming, however because the analytic software is able to pull in data from many sources
a detailed comparison of bit effectiveness in the different formation type can be carried out.
The Bit Performance allows the bit information to be filtered by:
● Vendor
● Size
● Formation
● Drives
● Mud types
● Run length (min and max)
Because the information from the major database providers is so detailed it is possible to analyse the
times taken for all operations for each of the wells. At the beginning of the process high level Key
Performance Indicator (KPIs) can be entered into the probabilistic planning software (see Figure 10), the
data analytic software calculates the information that is needed for the three data points of Best Case, Most
Likely and Worst Case. The definition of what operation is considered a KPI is customisable.

Figure 10 —Calculated KPIs

Provided the data is available the time analysis part of the data analytic software can be very detailed
allowing the detailed historical date to be used in the probabilistic planning software giving more accurate
cost and time analysis for projects. The data can be drilled into to look at background data for each KPIm
This is essential to understand the provenance of the results. This ability is useful for identifying wrongly
entered data. When incorrect data is identified it can be corrected in the source system and then gets
automatically fed back into the calculations.
IADC/SPE-180651-MS 9

Conclusion
The ability to bring information together from various sources and present it in a common graphical
format allows faster and more accurate engineering. In addition the ability to present information in a
format that is required by other software packages without further manipulation makes the use of that
information more likely in the planning process.
By having more information available from all relevant offset wells, we have experienced that the
planning is moving from previously being more generic to now being far more specific for the drilling
area.
It is possible to optimise well design and planned operations by actual experiences or historical data
without compromising operational safety nor quality.
As the standardised database provides a large amount of operational data the process of building a more
accurate probabilistic model has been improved.
To summarise we have experienced by managing data in a defined process that we not only save
engineering time, but we also manage in a high degree to optimise well design, leading to potentially
saving time and money. By planning with realistic data and using probabilistic methods for time and cost
estimates, we have also managed to produce far better AFE’s, with better matching to actual operational
time and cost, leading to better decision data for management.

Figure 11—Overview of Data Consolidation