IADC/SPE-178828-MS

Pipe Sticking Prediction Using LWD Real-Time Measurements

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

Joshua Hess, Baker Hughes Inc.

Copyright 2016, IADC/SPE Drilling Conference and Exhibition

This paper was prepared for presentation at the IADC/SPE Drilling Conference and Exhibition held in Fort Worth, Texas, USA, 1–3 March 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
In addition to formation evaluation, logging-while-drilling (LWD) measurements can also be used as a
real-time solution to issues that arise during drilling. For example, LWD measurements can be interpreted
to help determine if increasing mud weight is useful or if the rate of penetration (ROP) must be slowed.
What if measurements made with LWD tools can help predict whether or not a drill string will become
stuck while drilling? This paper describes a model that was created to predict situations with high risk of
the tool becoming stuck so they can be addressed during the planning and execution phases. The model
learns, or builds on each job experience, to improve future decisions.
Most experienced individuals involved in the drilling process understand the various ways that a drill
string can become stuck and factors that contribute to it becoming stuck. Many variables are attributed to
a bottom hole assembly (BHA) becoming stuck during drilling operations. This paper explains the
variables that were used in the model and why they were used. Some measurable variables are wellbore
characteristics, BHA characteristics, downhole pressure and temperature, drilling practices and mud
properties. These variables were gathered from 42 wells drilled on the Gulf of Mexico (GOM) shelf. More
specifically, the information was gathered from end-of-run reports, annular pressure (AP) logs, LWD
databases and surveys.
After the model was created, the model was tested against seven additional GOM shelf wells. The
model predicted correctly that in three of the seven wells, the drill string would be permanently stuck. In
another three of the seven, the drill string was temporarily stuck while the model claimed it would be
permanently stuck. In the final well, the model predicted that the drill string should be stuck, while it was
not; however, the client made decisions while drilling the section which indicated that they were fearful
of getting stuck.
If it can be shown that there is a risk of the drill string becoming stuck in a well, preventative as well
as after-the-fact measures can be used to adjust the mud system and other drilling parameters. Addition-
ally, fishing services can be prepared early or even moved to location earlier to cut down on non-
productive time (NPT). A model such as this provides an indication of the need to try preventive measures
to mitigate the situation before the problem occurs, or at a minimum provides additional information so
that the best course of action can be taken.
2 IADC/SPE-178828-MS

Background
Stuck pipe has a significant impact on drilling performance. A recent study (Muqeem et al. 2012)
attributed 25% of Non-Productive Time (NPT) to stuck pipe, at a significant cost to the drilling industry
This cost is often quoted as being in the hundreds of millions of dollars (see for example, Bradley et al.
1991, who estimated a cost of $250 million per year in the late 1980s). It is, therefore, of considerable
interest to develop methods of predicting when stuck pipe (both mechanical and differential) is likely to
occur. These methods may be developed on existing data sets and applied in real-time while drilling.

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

Studies using regression analysis for predicting drill string sticking extend back to at least the 1980’s.
In 1987, a large study conducted in the GOM on 131 cases of sticking concluded that multivariate
statistical analysis indicated the expected relationship between drilling parameters (Hempkins et al. 1987).
The study in this case included 28 surface measured or configuration (such as Drill collar OD) variables
and settled on a final 20 drilling parameters that were significant in predicting pipe sticking. They found
that their statistical models had a success rate of 80% to 87% in predicting pipe sticking.
During another study that was published in 1994, Belaskie et al. calculated friction factors in real-time
to warn the driller of possible sticking situations. They recommended monitoring three friction factors—
axial drag without rotation, axial drag with rotation and rotary friction—against operator entered
thresholds.
Later in Iran, researchers utilized a similar method to the one proposed in this paper by using regression
analysis to predict pipe sticking in a given area (Shoraka 2011). One primary difference between that
study and this paper is the use of LWD data in this study. Like the earlier 1987 study by Hempkins in the
GOM, the Shoraka study obtained an 80 to 86% success rate.
Over time, the focus on prediction has moved to real-time. This means that the variables used in these
prediction methods should be readily available at the wellsite (or at the remote monitoring center) while
drilling. For example, the friction factor method is an excellent solution if the operator has LWD tools
providing downhole torque and downhole weight on bit in the BHA; otherwise, the method is unachiev-
able.
Shoraka’s study in Iran utilizes formation pressure. In his research, he found that formation pressure
had the highest correlation with differential pressure (as he notes this is not surprising, because differential
pressure is the difference between hydrostatic and formation pressure), which he used to predict whether
or not a pipe sticking situation would occur. His approach is extremely effective if the formation pressure
is readily available for calculation. The variables used for the model in this paper are obtainable through
basic LWD services.
The method proposed in this paper included some variables that were not included in previous studies
such as: dogleg severity categories, drilling operator decisions, flow off pressure metrics, etc. As
technology improves items such as these and many more can be accurately captured in a database to allow
for quick, effective analysis to make real-time decisions and changes based on current observations.

Pipe Sticking Basics

This section briefly reviews the manner in which drilling strings become stuck in the borehole. Drill stings
can be stuck either mechanically or differentially.
Mechanical Sticking
There are three primary forms of being mechanically stuck: due to cuttings beds, key seating and shale
swelling.
Cuttings beds (Pack-Off)
Pack-off is a borehole effect where cuttings accumulate in a localized area causing the BHA to become
stuck. Pack-off is caused by cutting beds forming because of poor hole cleaning (Zhu 2011). There are
various factors that are associated with cleaning the hole properly. Some of the factors are related to the
IADC/SPE-178828-MS 3

mud properties and the fluid’s ability to suspend the cuttings and transport them to the surface. Other
contributing factors include the borehole trajectory, which may or may not slow down the progression of
the cuttings to surface. The rate of penetration and revolutions per minute (RPM) can affect borehole
cleaning because cuttings may be created faster than they can be removed from the hole. ROP and RPMs
can affect the rate at which cuttings are created and the size of cuttings created. An additional concern is
borehole quality; highly tortuous boreholes may have low spots in which cuttings beds can initially form.
Key Seating

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

Key seating is due to the drill pipe rubbing a ledge into the formation and causing the drill pipe to become
wedged. Factors that cause key seating are high doglegs where the drill pipe rubs against the formation
and poor formation integrity at the location of trajectory changes. Fig. 2 diagrammatically shows the
formation of a key seat.

Figure 1—Cuttings beds build up (Effendi 2012)

Figure 2—Key seating (Drilling Formulas, 2011)

4 IADC/SPE-178828-MS

Shale Swelling
Swelling of shales can cause the formation to close around the drill pipe lodging it in the hole. Various
mud properties such as chlorides and type of mud are usually attributed to the effect the mud has on
hydrating the shales and causing them to swell. Fig. 3 illustrates the effect.

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

Figure 3—Swelling Shales (Drilling Formulas 2011)

Differential Sticking
Differential pressure (Dupriest 2010) causes the drill pipe to behave as if suctioned to the wall of the hole,
and it is commonly caused by entering zones that have been depleted and are at a lower pressure than the
borehole. The BHA comes in contact with the borehole wall and the force created by the pressure
differential between the borehole and the reservoir where the tool is located is high enough to hold the tool
in place. A zone that has produced is often the biggest risk for differential sticking of the drill pipe. Fig.
4 diagrammatically shows the BHA being held against the formation by the overbalance of the wellbore
pressure compared to the formation pressure.

Figure 4 —Differential sticking (Wikipedia 2015)

Study Variables
The following variables were utilized in this study, and are separated into classes: those that were fixed
in value pre-run and those that were measured during the run. The values measured during the drilling run
were grouped into two classes depending on the possibility of controlling them in real-time.
IADC/SPE-178828-MS 5

Variables Available Pre-Run

Hole Size and Tool Size The borehole size, compared to the tool size, is a significant factor in tool
sticking because more stand-off is achieved with a larger difference between these two variables. The
larger difference affects the permissible flexibility of the BHA is in the hole.
Number and Spacing of Stabilizers The number and placement of stabilizers affects the BHA stand-off
from the wellbore wall. Stand-off cuts down on ledges that will be made when the BHA first penetrates
the formation. Additionally, the BHA is more stabilized to protect from undesired change in direction of

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

the well.
Reamer Reamers do not directly relate to tool sticking; however, they do have some characteristics that
are undesirable when it comes to tool sticking. One of these characteristics includes the effective flow
which is cut back at the bit and diverted to the reamer. The hole is cut with a different element (i.e. the
reamer) which can mean that cuttings are not the same as the ones created at the bit. The difference can
affect the cutting bed formation and removal of cuttings from the hole. Although, there is no direct link
with reamers causing tool sticking, it is a variable that needs to be considered.

Variables Available During Run (Uncontrolled)

The following variables utilized in this study were deemed as uncontrolled variables because they are not
changed at will during the bit run, while controlled variables can be altered as desired.
Measured Depth (MD) and True Vertical Depth (TVD) The measured depth and true vertical depth
do not directly affect the BHA’s chances of getting stuck. It is, however, understood that deeper wells tend
to have their own challenges. The measured depth and true vertical depth, in comparison with other
variables, can help to determine the general wellbore curvature in addition to potential problem spots if
drilling in the same area/field as other nearby problem wells.
Annular Pressure Sudden changes in annular pressure can indicate many sticking problems, such as
packing-off from shale swelling or cutting beds build-up. In essence, if the annular pressure increases
without a change in any other drilling metrics, then this could be an indication that there is some form of
blockage between the BHA and wellbore wall causing pressure seen at the LWD tool to increase.
Equivalent Circulating Density (ECD) The ECD is a good indication of proper borehole cleaning and
borehole integrity. The annular pressure changes proportionally with TVD the deeper you drill. Thus, a
fluctuation in ECD without a change in mud weight can indicate future downhole problems. Static mud
weight is the minimum ECD equivalent seen while the LWD tool is powered off during the last flow off
event. On the other hand, surge ECD is the maximum ECD equivalent seen while the LWD tool is
powered off during the last flow off event.
Temperature The temperature of the fluid and formation can have an impact on whether or not a tool
may get stuck. This is attributed to a change in the downhole environment that can lead to undesired
effects.
Inclination The inclination of the hole can affect the ability to clean the hole properly. Some inclinations
(typically near 45°) can be troublesome for borehole cleaning because cuttings tend to accumulate in this
area during circulation.
Dog Leg Severity (DLS) Dog leg severity is the curvature of the well typically denoted in degrees per
100 ft. High DLS can contribute to key seating if the drill pipe rubs into a formation that is soft.
Additionally, the higher the DLS, the more abrasive the rubbing is between pipe and the borehole wall.
6 IADC/SPE-178828-MS

Variables Available During Run (Controlled)

The remainder of the variables utilized in this study is deemed as controlled variables because they can
be altered as desired. These are often (but not always) referred to as operating parameters.
Rate of Penetration The ROP can play a role in tool sticking because at higher ROP cuttings are created
at a quicker rate than they can be removed from the hole and can cause a pack-off.
Back Reaming Back reaming is a good drilling practice because it helps to bring cuttings out of the hole.
Cutting beds are reduced by back reaming multiple times which provides more time for cuttings to come

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

to surface before new cuttings beds are created.
Rotary Speed Rotary speed is a parameter that describes the rotation rate of the drill string and bit, and
controls the frequency of contact of the BHA with the formation. Rotary speed may play a key role in the
creation of formation cuttings. Some cuttings are easier to remove from the hole than others. Additionally,
rotation of the drill string helps to stir up cuttings downhole to further assist with borehole cleaning.
Therefore, rotary speed can be a factor that contributes to borehole cleaning and formation of cutting beds.
Flow Rate The effective downhole flow rate can play an integral role in proper borehole cleaning. The
faster the fluid is circulated, the more likely the cuttings can be transported out of the borehole. Flow rate
can significantly cut down on the formation of cutting beds.
Mud Weight (MW) One of the mud weight’s primary functions is to hold the borehole open while not
fracturing the formation. If the pipe is unable to move but the annular pressure is constant, this may be
an indication of differential sticking. When a depleted zone is penetrated, the pressure differential between
the mud column and the formation can be great enough to hold the pipe stuck in place against the
formation wall.
Plastic Viscosity (PV) and Yield Point (YP) Plastic viscosity is a mud property that indicates a drilling
fluids ability to bring cuttings out of the hole while maintaining a suitable ROP. Yield point is a measure
of the ability to suspend cuttings in the drilling fluid, so that cuttings can be removed from the hole. Plastic
viscosity and yield point indicate the ability of the drilling to remove cuttings from the hole, and the rate
at which drilling can be optimized.
Gel Strength Gel strengths (Gels) are a measure of the force necessary to cause the mud to flow again
after a period of non-circulation. They are, therefore, an indicator of the cuttings suspension properties of
a drilling fluid. The gels are a measurement of the shear stress measured at low shear rate, taken after
different mud-setting intervals. The gel measurements used in this model are measured after 10 second
intervals and 10 minute intervals, hence 10 sec gel and 10 min gel.
Chlorides Chlorides, the amount of salts in the drilling fluid (NaCl and KCl), is one of the best ways for
controlling shale swelling while drilling.
Oil/Water Ratio The oil-to-water ratio indicates the ratio of oil to water in the drilling fluid. Oil or
synthetic-based mud is desired over water based mud to cut down on the chances of tool sticking in
addition to providing better hole cleaning and lubricating properties.

Additional Metrics for Analysis

By creating new variables, data already present is used to emphasize significant attributes that would not
be captured otherwise. Derived variables used in this study were as follows: 1. Hole Size – Tool Size, 2.
Surge ECD – Static Mud Weight, 3. Surge ECD – ECD, 4. ECD – Static Mud Weight, 5. Static Mud
Weight– MW, 6. ECD – MW, 7. Ln(Temperature), 8. MD – TVD, 9. Ln(PV), 10. Ln (YP), 11. PV ⫹ YP,
and 12. 10 min gel – 10 second gel. These variables were denoted separately to set them apart from
measured variables.
IADC/SPE-178828-MS 7

Analysis/Results
After all of the data was collected and the extra variables were created, the information was entered into
statistical modelling software (linear regression software). Fig. 5 shows the coefficients calculated by the
software. The first and third columns are the coefficient values that are multiplied with the measured
values from the job. All forty items are summed to receive a value that is close to 0, implying the tool will
likely not get stuck, or 1, implying the tool will likely get stuck (in other words, a binary result).
Whichever the sum is closest to is the prediction. Additionally, the model had R2 values of .902 at a

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

confidence level of 95%. The R2 value is a measure describing the fit between the data and the model
where a value of 1.0 is a perfect fit. Fig. 7 shows an example of the statistical model from a well drilled
on the Gulf of Mexico (GOM) shelf in May of 2014. The equation below depicts the variables and how
they are applied:
(1)

Figure 5—Model Equation Coefficients and Associated Variable

Figure 6 —Variables Removed from the Model

8 IADC/SPE-178828-MS

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

Figure 7—Example of Model Calculations from GOM Shelf Well in May 2014

The y-intercept is b0 and is added to the sum of the coefficients multiplied by their measured variable.
The coefficients b1 to b39 represent the coefficient from the statistical modelling, based on approximate
ranges for the variables entered into the model, as well as weights of the variables. Ranges and weights
IADC/SPE-178828-MS 9

for given variables are show in Fig. 8. The coefficients are first normalized to compare items evenly and
then assigned weights based on how the items contributed to the event of becoming stuck or not stuck.
The data in Fig. 7 for a GOM well indicates that the BHA will become stuck in the hole (the equation
derived a value close 1.0). This BHA did become permanently stuck in the hole and was abandoned.

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

Figure 8 —Ranges/Weighting Recorded from Model amongst the 42 Wells

10 IADC/SPE-178828-MS

Variables Removed
Not all of the values that were collected and derived were used. Fig. 5 shows all the values that were used
in the calculation. Fig. 6 shows all of the values that were measured and derived but were not significant
when creating the model. That does not mean that these variables were not important; rather, they were
not significant across the wells that were used to create the model.

Limitations of the Model

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

The model does have some limitations as to what can be entered in it. First, the model is comprised solely
of wells drilled on the outer continental shelf in the Gulf of Mexico (GOM). Therefore, it gives a reliable
prediction for these types of wells. US land or other jobs cannot be entered into the model because drilling
practices vary from place to place. Also, the model cannot manage jobs that have zero or one stabilizer
only in the BHA because the minimum and maximum spacing that must be entered throws the model off
when using an extremely high null value (which is an accurate depiction of such a BHA but fails in the
model). In other words, the model did not use slick BHAs, or single stabilizer BHAs.

Application
Based on the created equation, an Excel sheet that automatically calculates the probability of sticking was
created. The user must enter the values collected from LWD measurements, in addition to the well name.
Fig. 9 shows an example of this sheet.
IADC/SPE-178828-MS 11

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

Figure 9 —Sheet Created to Enter Metrics and Auto Calculate Prediction

Future Models
42 shelf GOM wells were used to build the model equation, and an additional 7 wells were used to
independently test the model. The 7 wells still fit the criteria of shelf GOM wells with two or more
12 IADC/SPE-178828-MS

stabilizers in the BHA. Of the seven wells, three were accurately predicted. The remaining four were
closely predicted: the model indicated the BHA would get ⬙stuck permanently⬙ whereas the BHAs were
⬙temporarily stuck⬙ in three situations, with the client taking actions on the fourth to which indicated that
it ⬙may get stuck.⬙
Fig. 10 and Fig. 11 are charts that were made to show how the model equation changed by
incrementally adding wells to reform the model. The model is recreated in the same fashion with more
information.

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

Figure 10 —Table for Model Equation Improving Prediction as More Wells are Added
IADC/SPE-178828-MS 13

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

Figure 11—Table Depicting Variable Weighting Changes in Re-derived Equations as Wells are Incrementally Added to Re-derive the
Equations

Fig. 10 shows that as new wells are added, the equation evolves to accurately predict a larger
percentage of wells. The columns are filled with the number of wells used to build the new model. The
rows are the actual outcome of the well (1 indicates the BHA was stuck / 0 indicates the BHA was not
stuck). The cells marked in red are indications that the new equation inaccurately predicted the outcome
for whether or not the BHA became stuck. At the bottom of the chart, the ⬙model wrong⬙ indicates the well
information was used to make the new model equation and was inaccurately predicted, while ⬙independent
14 IADC/SPE-178828-MS

wrong⬙ indicates that the well information was not used to make the model equation and was inaccurately
predicted.
Fig. 11 shows the percentage change in variable coefficients from incrementally adding a well at a time.
The columns show the changes in coefficients from well xx to well yy. In the first column, the percentage
of change for a given variable between an equation derived using measurements from 42 wells and an
equation derived using measurements from 43 wells are shown. Fig. 11 was added to demonstrate the
benefits of gathering more wells in a given area to continuously re-derive an equation to explain the
relationship amongst variables in a given area. The rows are the variables for which the percentage

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023

changes were made. The very bottom shows the total percentage change.

Summary
The modelling methodology follows the logic of a determined person gaining more experience over time
to better predict what the outcome of a situation may be. The variables selected and used incorporate most
of the variables that are not only attributed to tool sticking but also are monitored on an individual basis
to make changes as necessary throughout the drilling process. Risk assessment for a well is typically
performed at specific stages as part of the well planning process rather than continually throughout the
process until the well is drilled. In a study by Saudi Aramco, stuck pipe events account for 25% of annual
NPT (Muqeem 2012), so any tools to reduce or mitigate stuck pipe events will reduce this annual
percentage that stuck pipe contributes to yearly NPT.

Conclusions
1. There is sufficient data available to accurately predict sticking events using linear regression
models. These models can be applied in real-time, as the well is being drilled.
2. Linear regression equations will improve prediction accuracy as more wells are added to the
model.
3. Better decisions can be made using the model with regard to adding preventative fluids to the mud
system, ordering and having fluid additives available in an emergency to eliminate NPT, or
ordering and having fishing tools available on location.

References
Belaskie, J.P., McCann, D.P., Leshikar, J.F., 1994, A Practical Method to Minimize Stuck Pipe Integrating Surface and
MWD Measurements, presented at the 1994 IADC/SPE Drilling Conference, Dallas, Texas, 15-18 February 1994,
SPE-27494-MS.
Bradley, W. B., Jarman, D., Plott, R. S., Wood, R. D., Schofield, T. R., Auflick, R. A., & Cocking, D, 1991, A Task Force
Approach to Reducing Stuck Pipe Costs, SPE-21999-MS
Drilling Formulas, 2011, Key Seat Causes Stuck Pipe, http://www.drillingformulas.com/key-seat-causes-stuck-pipe/ (ac-
cessed 8 October 2015).
Drilling Formulas, 2011, Shale Instability Causes Stuck Pipe, http://www.drillingformulas.com/shale-instability-causes-
stuck-pipe/ (accessed 8 October 2015).
Dupriest, F.E., Elks, B. & Ottesen, S. (2010, January 1). Design Methodology and Operational Practices Eliminate
Differential Sticking, SPE-128129-MS.
Effendi, H.S., 2012, Petroleum Support, Mechanical Sticking – Mechanism of Stuck Pipe, http://petroleumsupport.com/
mechanical-sticking-mechanism-of-stuck-pipe/ (accessed 8 October 2015).
Hempkins, W.B., Kingsborough, R.H., Lohec, W.E., Nini, C.J., 1987, Multivariate Statistical Analysis of Stuck Drillpipe
Situations, SPE-14181-PA.
Muqeem, Muhammad A., Weekse, Alexander E., Al-Hajji, Ali A., 2012, Stuck Pipe Best Practices – A Challenging
Approach to Reducing Stuck Pipe Costs. Presented at the SPE Saudi Arabia Section Technical Symposium and
Exhibition, Al-Khobar, Saudi Arabia, 8-11 April 2012, SPE-160845-MS.
IADC/SPE-178828-MS 15

Shoraka, S.A.R., Shadizadeh, S.R., Pordel Shahri, M., 2011, Prediction of Stuck Pipe in Iranian South Oil Fields Using
Multivariate Statistical Analysis, Presented at the Nigerian Annual International Conference and Exhibition, Abuja,
Nigeria, 30 July – 2 August 2011, SPE-151076-MS.
Wikipedia, 2015, Differential Sticking (29 June 2014 revision), https://en.wikipedia.org/wiki/Differential_sticking (ac-
cessed 08 October 2015).
Zhu, X., Liu, S., & Tong, H., 2011, Research on Transport of Cuttings in Gas Drilling Horizontal Well and Cuttings
Falling-prevent Joint Development, SPE-145612-MS

Downloaded from http://onepetro.org/SPEDC/proceedings-pdf/16DC/3-16DC/D031S026R008/1390342/spe-178828-ms.pdf by U. of Alberta Library user on 15 September 2023