IPTC 10709

Physics-Based Well DesignBeyond the Learning Curve

B. Kline, SPE, K. Chandler, SPE, S. Keller, SPE, S. Ottesen, SPE, V. Gupta, SPE, and M. Tenny, ExxonMobil Upstream
Research Co.

This paper was prepared for presentation at the International Petroleum Technology
Conference held in Doha, Qatar, 2123 November 2005.
This paper was selected for presentation by an IPTC Programme Committee following review
of information contained in an proposal submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the International Petroleum Technology Conference
and are subject to correction by the author(s). The material, as presented, does not
necessarily reflect any position of the International Petroleum Technology Conference, its
officers, or members. Papers presented at IPTC are subject to publication review by Sponsor
Society Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this
paper for commercial purposes without the written consent of the International Petroleum
Technology Conference 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 where and by whom the paper was presented. Write
Librarian, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract
In a drilling program, the design of wells has historically been
an incremental process in which compiled experience from
previously drilled wells is the principal driver for realizing
drilling design. The efficiency of drilling in a field, and as a
result, the cost, typically improves significantly initially,
before reaching a point at which little improvement is
achieved. The progression from the initial well to the final
wells of a mature drilling program is often referred to as the
"drilling learning curve." Rather than focus on getting the
design right after several wells, it is possible to outperform the
learning curve by developing and applying fundamental
physics-based models and computational optimization engines
up front to quantitatively analyze well design options. These
proprietary tools have been applied extensively in field
development planning, as well as in real-time, to ensure
optimal drilling performance.
This paper describes
ExxonMobil's approach to physics-based modeling, advanced
mathematical optimization, and real-time updating to achieve
drilling performance that both accelerates and undercuts the
historical learning curve.
Specific business examples
illustrating successful applications of this approach are
presented.
Introduction
Optimal well design requires attention to both fundamental
engineering principles and site-specific learnings. The drilling
learning curve, which identifies adjustments to initial design
parameters to improve overall well design, can frequently cut
50% or more from well costs over the course of a continuous
program. While a variety of learning-based well design
techniques have served the industry well, the biggest prize is
achieved by utilizing physics and mathematics to shift the
optimization process from the learning curve to the initial well
design getting it right the first time.
ExxonMobil's approach to going beyond the learning curve
is based on routine application of proprietary technology that

couples fundamental physics-based models with statistical

analyses. The models enable the accurate prediction of
expected drilling performance over a wide range of
operational conditions.
By accounting for downhole
parameter uncertainty, engineers effectively optimize this
performance over all expected scenarios. While drilling,
collected real-time information aids in reducing these
uncertainties, enabling engineers to revisit the well design,
model and optimize the rest of the well, and adjust operations
accordingly. These tools may all be leveraged through inhouse visualization capabilities that enable the efficient
comparison of competing well designs and the evaluation of
planned and current drilling operations. The resulting well
plans are robust to deviations in expected downhole
parameters, resulting in wells that are drilled quickly and
inexpensively while eliminating hole-quality related nonproductive time incidents, even on the first well of a drilling
program.

Cost

Copyright 2005, International Petroleum Technology Conference

Historical approach

Physics-Based Well Design

Well #
Figure 1: Representation of the savings of the Physics-Based
Well Design

Figure 1 is a graphical representation of the benefits of the

Physics-Based Well Design approach. First, it is important to
note from this figure that the cost for the first well for the
Physics-Based Well Design is lower than the historical
approach, which means that this approach is beneficial even
for exploration wells. Increasingly, fields are being developed
with fewer, high-cost, high-rate wells, where the learning
curve does not have several wells over which to improve
significantly. For this reason, it is also important to note that
for the Physics-Based Well Design, the individual well costs
approach the final minimum well cost more rapidly than in the
historical approach, meaning that significant cost savings can

IPTC 10709

be achieved without drilling many wells. Finally, the

minimum achievable cost for drilling using the Physics-Based
Well Design is lower because it is able to deliver a more
thorough evaluation of possible improvements than was
previously possible. The savings from this approach,
therefore, are captured continuously throughout the entire
drilling program.
In this paper, ExxonMobil's specific distinguishing drilling
technologies are described and selected recent examples of
how these technologies have been applied worldwide are
presented.
Physics-based modeling
ExxonMobil's Integrated Hole Quality (IHQ) technology is an
integrated analysis of wellbore stability, hole cleaning,
hydraulics, lost circulation, well control, stuck pipe, torque &
drag, drilling fluid design, wellbore architecture, drill string
design, and operating practices that delivers recommendations
that optimize drilling procedures and rig capabilities. This
technology is based on fundamental physics, experimental and
field results, and engineering best practices. IHQ relies on
certain downhle parameters that have uncertain values. As
presented in a later section, the IHQ technology can account
for these uncertainties through Quantitative Risk Analysis
(QRA), which uses mathematics-based analysis of trouble cost
mechanisms, well cost elements, and statistics to manage
unknowns. Before discussing this aspect of the well design
approach, it is important to present first the individual IHQ
models.
10250

10500

10750

Depth (ft MD)

11000

11250

11500

11750

12000
Predicted Caliper (10.4% B/O)

12250

Actual Caliper (10.9% B/O)

12500
12.0

12.5

13.0

13.5

14.0

14.5

Hole Diameter (inch)

Figure 2: Caliper prediction for an offset well

The wellbore stability model identifies the minimum

drilling fluid density that avoids hole collapse as well as
calculates the anticipated hole caliper for mud weights below
the collapse mud weight. The model was developed from

first-principles physics-based stress modeling based on earth

stresses, pore pressure, well architecture, fluid type, and rock
strength of the near-wellbore region and then verified using
field data. The rock strength is correlated to the shale surface
area obtained through measuring the rock dielectric constant
from cuttings retrieved from offset wells or from the current
well in real time by personnel on the drilling rig [1]. By
performing this analysis while drilling, the mud weight can be
adjusted to prevent wellbore instability before it occurs.
Figure 2 shows a typical output of the wellbore stability
model, comparing the model's predicted caliper to the
measured caliper from the same well after it has been drilled
and logged.
The purpose of the hole cleaning model is to specify
drilling operational parameters (e.g., mud flow rate and
rheology) and guidelines (e.g., drill string tripping speed) that
are particularly important in high-angle and extended-reach
wells. The model was developed based on physical principles
of cuttings transport and deposition and was verified
experimentally. This model is used to prevent stuck pipe from
poor hole cleaning, excessive drag from cuttings buildup, and
as a component for predicting equivalent circulating density
(ECD) effects of a cuttings bed or suspended cuttings.
Wellbore hydraulics are modeled incorporating the effects
of not only drill cuttings, but also temperature-dependent
density and rheology of the drilling fluid, which are critical
phenomena for non-aqueous drilling fluids. Using a transient
thermal simulator to model viscous heating from fluid
circulation and heat transfer between the well and its
surroundings, the standpipe pressure capacity required by the
circulating system can be designed to guarantee that the rig
pumps and equipment can handle the proposed drilling
operations. This model also predicts the pressure from the
circulation of the drilling fluid, which must be maintained
below the local fracture pressure to prevent lost returns. The
hydraulics model predictions have been verified with recent
field tests.
In the event that lost returns are experienced, models have
been developed to strengthen the wellbore and mitigate further
fluid loss using fracture-closure stress technology [2]. The
guidelines are the result of fundamental modeling of the nearwellbore stresses by exploring the fracture mechanics of the
system. This model has resulted in a process to control lost
returns by mechanically enhancing the fracture gradient in the
vicinity of a hydraulically induced fracture.
The differentially stuck pipe model is used to determine
the risk of stuck drill pipe, casing, or wireline tools given
bottomhole apparatus (BHA) design, casing centralization,
well design, and rig and equipment capacities. This model
was developed from experiments using a proprietary stuck
pipe apparatus, as well as computationally simulated cases
using finite-element modeling. When coupled with a torque
and drag model, recommendations on BHA design and casing
centralization in addition to well path improvements can be
made.
Each of these physics-based models, when used
independently, can significantly improve a well design. The
recommendations from each of these models strongly interact,
and may even conflict with one another. For instance,
consider a well in which optimizing the mud weight for

IPTC 10709

wellbore stability may result in lost returns. Without taking

both aspects of the situation into account, it becomes difficult
to design this well efficiently. Now imagine trying to balance
multiple parameters according to the results of several
analyses the process quickly becomes daunting. For this
reason, the individual models are combined under a single,
unified framework called Integrated Hole Quality (IHQ). IHQ
acts as a single model that accounts for all aspects of hole
quality simultaneously. Through IHQ, it is possible to achieve
a well design that avoids hole-related problems, given that the
downhole earth parameters, such as rock strength and stresses,
are known exactly. Of course, precisely knowing these
parameters is not possible without having drilled the well first,
so it is important to account for uncertainties in the well
design.
Dealing with uncertainties
The individual IHQ models are integrated using a
mathematical technique called Quantitative Risk Analysis
(QRA) that incorporates uncertainty in data and calculates a
probability of avoiding hole-related trouble for a well design.
Optimizing the well design using the QRA approach results in
what is referred to as the Physics-Based Well Design, which
is considered a proprietary ExxonMobil technology.
The QRA method determines the statistical probability of
encountering hole-quality related problems based on the IHQ
technology [3]. Consider the previous example in which one
is designing a well that may encounter lost returns and/or
wellbore instability, only now the rock stress downhole is not
well known. The QRA approach balances these two effects
under quantified uncertainty to deliver a design that is robust
to both problems by maximizing the probability of avoiding
both hole quality-related problems simultaneously. Figure 3
demonstrates the result of a typical QRA analysis. In this
graph, the optimal drilling parameters correspond to the point
at the peak of the graph's surface.

0.9
0.8
0.7

5 strings (high initial cost)

0.6
0.5
0.4
0.3
0.2
0.1

Distribution

ss
Probability of Drilling Succe

4 strings

The QRA technology enables the engineer to plan complex

wells that address several design challenges under uncertainty
at once. For instance, mud weight, fluid rheology, drilling
rate, circulation rate, rotation rate, rig capacity, and casing
strategy, can be optimized to guarantee wellbore stability,
proper hole cleaning, lack of lost returns and/or well control
incidents, and avoidance of differential pressure sticking. The
resulting Physics-Based Well Design allows for greatly
accelerated learning curves, yielding record extended reach
wells without any hole-related trouble and even wells that
could not otherwise be drilled. In essence, this technology
enables the driller to keep the hole that is drilled.
Real-time data acquisition
Planning wells under large uncertainties can lead to overly
conservative designs. For this reason, real-time information is
obtained from wells while they are drilled to revise their
designs given the reduced uncertainty, thereby adapting to
new information as it is received.
For instance, downhole pressure while drilling (PWD)
tools have been used to measure the ECD at the bit to prevent
lost returns events. The measured pressure is used not only to
verify in-house computational models, but also to determine
the true fracture gradient downhole, narrowing modeled
uncertainties. This information is then used to model lost
returns and wellbore stability more accurately, potentially
leading to changes in design.
Similarly, temperature measurements allow more accurate
modeling of the hydraulics of drilling operations. The
temperature affects the standpipe pressure and ECD, which are
modeled using the previously discussed methods. As a result
of revisiting the models with these new measurements, the
engineer may choose to adjust the mud rheology or density.
This type of feedback enables the avoidance of lost returns
episodes before they occur.
Real-time measurements are not limited to downhole tools,
either. Drill cuttings are collected as they are carried over the
shakers and correlated to their depth of origin. The cuttings
are measured using dielectric constant measurement (DCM) to
determine the rock strength at that depth. This information
yields trends that may result in a new wellbore stability
prediction and consequently a change in recommended mud
weight, preventing hole instability before it is observed.
As new data are collected, more precise information is
incorporated into the models, and hole-related non-productive
time can be avoided before potentially costly symptoms such
as lost returns, tight hole, or stuck pipe are experienced. This
technology results in significant savings while drilling.
Visualization

450

Well
Cost

0
17.5

410
370

330
17.0

Expected Well Costs

16.5
Mud Weight
(ppg)

250

16.0

Figure 3: The output for a QRA analysis

15.5

290

Flow Rate
(gpm)

Another technology that ExxonMobil employs to assist in

Physics-Based Well Design is its state-of-the-art visualization
capabilities. The visualization facilities range from desktop
visualization to fully immersive three-dimensional
environments that enable collaboration among groups of
engineers and geoscientists. The visualization of a planned
well or field typically begins in the early planning stages of
the field development, but may be revisited during the drilling
and production of the field.

The environments display formations, well paths, hole

architecture, log data, and analysis output from the well design
tools. Additionally, this environment enables the storage and
extraction of data related to, but not limited to, earth stresses,
rock strength, and pore pressure. The data may come from a
combination of offset wells and other analyses, such as the
integrated pore pressure prediction technology.

Figure 4: A collaborative visualization session

IPTC 10709

resulting design was planned with a 90% chance of avoiding

hole-quality problems and was successfully drilled with zero
hole-related NPT.
Another example of the benefits of the Physics-Based Well
Design is the drilling performance in the Chayvo field, on
Sakhalin Island on the east coast of Russia [5]. The surface
locations for the development wells are approximately 6-7
miles (10-15 km) from the offshore locations of the
exploration and appraisal wells which were drilled as straight
holes. The onshore development resulted in significant
savings versus an offshore platform design. The first four
wells represent record wells for ExxonMobil and are, to
knowledge, among the ten longest extended reach wells in the
world. Figure 5 presents the profile for one such well. These
wells were designed by applying IHQ/QRA and learnings
from global extended-reach-drilling (ERD) experiences.
Cuttings were collected and tested during drilling to update the
wellbore stability model to verify that the proper mud weight
was being used. Another key concern for this development
was ECD management. The ECD was carefully monitored
using PWD tools to prevent lost returns while drilling these
challenging horizontal wells. As a result, these wells were
drilled successfully with minimal hole-related problems.
ERD Land Rig

Field Examples
There are numerous examples worldwide of how ExxonMobil
has used the Physics-Based Well Design, covering six
continents, on land, in shallow and deep water, with vertical
and highly deviated wells, and in almost every type of weather
climate and operating condition imaginable. Here, three such
cases are presented Hibernia, Chayvo, and a development
drilling program in the Middle East.
In 2003, a well was drilled from the Hibernia platform, off
the eastern coast of Newfoundland [4]. The well was highly
deviated, and planned to have the world record longest reach
of 30,699 ft MD (9400 m) for its vertical depth (12,993 ft or
>3000 m TVD). The well design relied on proprietary models
to achieve substantial life cycle cost savings over a subsea
well design. Real-time drill cuttings analyses using DCM
were performed to monitor wellbore stability. As a result, the
original recommendations were revised to incorporate a
lowered mud weight requirement in the 8-1/2-inch hole. The

Chayvo Bay
Barrier Island

Sea of Okhotsk

18 5/8'' Casing

1000
Depth TVD (m)

To determine how to access the reserves of a field, a multidisciplinary well team may view the geologic data in the
three-dimensional visualization environment (Figure 4).
Geoscientists and engineers develop quantitative forecasts of
downhole conditions, such as pressure, temperature, and stress
gradients. They can easily extract data for their analyses along
any proposed well path to assist in well planning. The data is
communicated by special linkages to analysis software, such
as those described in the context of the Physics-Based Well
Design. Using the recommendations of the well design tools,
this team can then weigh well path options versus the target
requirements and drilling location options. In this way, using
customized visualization software, they collectively and
interactively choose the best wellpath to reach the reservoir
targets. Through this process, visualization capabilities enable
the efficient evaluation of competing well designs as well as
the rapid analysis of planned and current drilling operations.

Hold Angle 76.68

13 5/8'' Casing

2000

Hold Angle 90.00

9 5/8 Casing

3000

7 Liner
10423 m MD
2600 m TVD

4000
-1000

1000

2000

3000

4000

5000

6000

7000

8000

9000

Vertical Section (m)

Figure 5: An extended reach well in the Chayvo field

For the drilling program in the Middle East, QRA

combined with economics was used to make a drilling design
decision. A balance between managing wellbore instability
and differential pressure sticking was required to drill a hole
interval. The QRA analysis showed that the probability of
success was higher if an additional string of casing were run.
However, an economic analysis of the results indicated that
managing the uncertainty by not running the casing string was
justified by savings of $1.5 million per well. Figure 6 shows
the recommended mud weight, flow rate, and rotary speed for
the studied interval. These wells were drilled successfully at
lower cost due to the coupling of physics-based modeling and
statistical analysis.

IPTC 10709

100%
90%

Probability of Success (%)

80%
70%
60%
50%
40%
Stable Wellbore

30%

430 GPM / 120 RPM

500 GPM / 120 RPM

20%

550 GPM / 120 RPM

10%

BHA w/ 270' DC (NFR 1-4))

BHA w/ 220' DC (Planned)

0%
11

11.5

12

12.5

13

13.5

14

14.5

Drilling Fluid Density (ppg)

Figure 6: QRA results from a Middle East drilling project

These examples are but a few of several success stories

worldwide in which ExxonMobil has applied the PhysicsBased Well Design principle to drill the most cost effective
wells possible. The historical approach of the drilling learning
curve is no longer the method of choice. Experience shows
that through understanding of the fundamentals, application of
statistics and optimization, real-time data collection and use,
and three-dimensional visualization, the learning curve can be
outperformed to yield significant benefits, no matter how
many wells are to be drilled.
References
1. Steiger, R. P., and Leung, P. K.: "Quantitative Determination of
the Mechanical Properties of Shales," paper SPE 18024
presented at the Annual Technical Conference and Exhibition of
SPE, Houston, 5-8 Oct. 1988.
2. Dupriest, F. E.: "Fracture Closure Stress (FCS) and Lost Returns
Practices," paper SPE/IADC 92192 presented at the SPE/IADC
Drilling Conference, Amsterdam, 23-25 Feb. 2005.
3. Ottesen, S., Zheng, R. H., and McCann, R. C.: "Borehole Stability
Assessment Using Quantitative Risk Analysis," paper
SPE/IADC 52864 presented at the SPE/IADC Drilling
Conference, Amsterdam, 9-11 Mar. 1999.
4. Elsborg, C. C., Power, A. H., and Schuberth, P. C.: "Hibernia
Record Well Breaks Extended Reach Drilling and Completion
Envelope," paper SPE/IADC 92347 presented at the SPE/IADC
Drilling Conference, Amsterdam, 23-25 Feb. 2005.
5. McDermott, J. R., Viktorin, R. A., Schamp, J. H., Barrera M. W.,
and Fleming, J. M., Keller, S. R.: "Extended Reach Drilling
(ERD) Technology Enables Economical Development of
Remote Offshore Field in Russia," paper SPE/IADC 92783
presented at the SPE/IADC Drilling Conference, Amsterdam,
23-25 Feb. 2005.