Vous êtes sur la page 1sur 7

SPE 94798

Drill-Bit Catalog and Bit Index: A New Method for Bit Performance Evaluation
P. Macini, SPE, and M. Magagni, SPE, U. of Bologna, Pietro Valente, Eni E&P Div.
Copyright 2005, Society of Petroleum Engineers
This paper was prepared for presentation at the SPE Latin American and Caribbean
Petroleum Engineering Conference held in Rio de Janeiro, Brazil, 20 23 June 2005.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in a proposal submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to a proposal of not more than 300
words; illustrations may not be copied. The proposal must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract
The paper analyzes the application of a new method for the
evaluation of bit performances. At present, the most utilized
methods for bit performance evaluation are mainly based on
the cost-per-foot. This approach is definitely a good reference
system, but it represents only a mere economic calculation,
and is not well suited for keeping into account drilling
difficulties due to formation variability and directional drilling
conditions. The objective of the research is both the analysis
of a new drilling model for bit performance evaluation, known
as Bit Index (BI), and the investigation of the possibility of BI
to establish sound guidelines for the development of a
mathematical tool able to help during well planning, well
drilling and post-well analysis. The model was tested by
mining data from the Drill-Bit Catalog (DBC) prepared at the
Exploration & Production Division of Eni Group (Italy). The
DBC is a database that contains both parameters for the
geotechnical/geomechanical characterization of the formations
and bit record/performance data. Within the DBC, three wells
have been identified, recently drilled in Southern Italy. These
wells have been used for the comparison and evaluation of bit
performances through the BI methodology. Main results
allowed to identify the bits (labeled with IADC code) that
drilled with the best performances in the above three wells.
The proposed model, although in its early stage of
development, proved to be economic and reliable. The quality
of bit performance analysis obtained with this model seems
also more effective than the traditional quick look analysis,
performed on the bit records or on the pure evaluation of the
cost-per-foot.
Introduction
The drilling industry has experienced since long time an
impressive technological advancement in drilling bit design
and performance, in order to meet the continuously changing
and more demanding needs of Operators. Today, longer time

on bottom and faster penetration rate are an increasingly


demanding issue for drilling bits. To this purpose, the
technology of drilling bits has evolved, and continues to
advance to meet the challenging needs of todays drilling
industry. Bit performance evaluation and bit selection is of
paramount importance for the drilling engineer. Many
attempts to model and/or predict drill bit selection and to
evaluate bit performance have been developed for a number of
years. As advancements in drilling bit technology and drilling
measurement system have increased, so have the number and
the sophistication of evaluative models available. Almost all
the approaches are similar in objective, to make drilling bit
selection less erroneous and more logical1. Recently, the
methods for the evaluation of bit performance improvements
have not always kept the pace with technological
advancements. Methods such as offset wells and cost per foot,
specific energy, bit factor, formation drillability, design index
and neural networks have been used for bit performance
evaluation2-8, although most of them still seems not
completely satisfactory.
The first studies on bit performance prediction and
evaluation were based on the modeling of the wear rate of
both cutting structure and bearings of milled-tooth bits9.
Despite of this, drilling engineers have selected bits mainly on
the guidelines of the best performer in a particular area, or
on performances of similar bits in offset wells. However, data
from offset wells contain a complex relationship between
drilling parameters, well path, formation properties, drilling
fluid properties, hydraulics, and bit design, limiting this
approach10-12. Bit performance evaluation and bit selection
carried out by means of the cost-per-foot is a good reference
system, but it represents only a mere economic calculation,
and is not well suited for keeping into account drilling
difficulties due to formation variability and directional drilling
conditions. Albeit one of the most accepted method, today its
effectiveness has drastically been reduced by the application
of modern drilling technologies, more demanding in term of
well path control and sophisticated downhole tools2.
The specific energy approach has been used to investigate
different bit types, different designs of the same type, and
modifications to the same design. Specific energy is defined as
the energy required for removing a unit volume of rock. It is a
measure of bit performance and is directly compatible with
methods based on the cost per foot or on offset wells2, 13.
Drilling models based on formation drillability investigates
the drilling response of a particular type of formation to
different types of bits and cutters. Although these methods
serve as an effective means for bit type selection, they does

not always establish the required correlation between


formation changes and the associated need for different bit
types14-16.
The approach based on design index is a methodology
developed for PDC bit selection, then expanded to include
roller cones bits. This index takes into account bit
characteristics such as the bit profile, number of blades and
cutters, cutter size, junk slot area and back rake. The design
index is based on determining the ideal bit for a given set of
drilling conditions, and then matching it with the bits proposed
by manufacturers17.
Recently, neural networks have been used to identify the
complex relationships between drilling parameters, well path,
formation properties, drilling fluid properties, hydraulics, and
bit design, when sufficient data exists. These can be supplied
by the increasingly sophisticated downhole measuring tools
that can provide large data sets with many recorded variables
related to drilling process. In this case, different neural
networks can be designed for a region or a field to predict
unknown parameters10.
However, the parameters affecting bit performances often
have a complicated pattern. Data from offset wells contain
complex relationships between drilling and operational
parameters, well path, formation type and its geomechanical
behavior, drilling fluid properties, hydraulics, and bit design,
which are not yet completely understood10, 18-21. Moreover,
most of the above approaches do not have a consistent
definition for the term performance, which complicates their
use and comparison2. In the light of the above, this research is
aimed to identify a different indicator of bit performances, in
the attempt to make more objective the process of bit
performance evaluation and, thus, of bit selection.
Bit Index
The objective of this research is the analysis, implementation
and application of a drilling model for bit performance
evaluation named Bit Index (BI). In this model, paramount is
the investigation of the possibility of BI methodology to
establish guidelines for the development of a mathematical
tool able to help during well planning, drilling and post-well
analysis. The idea of BI is clearly derived from the research of
Perrin, Mensa-Wilmot, and Alexander2, in the attempt to make
this method somewhat simpler and applicable in practical
cases with data available during drilling operations. The BI
methodology utilizes the simple data normally recorded while
drilling, collected from bit records, master logs and, if
necessary, from wireline logs. It distinguishes between bits
independent and dependent operational parameters, and
establishes the cost per foot as an indirect evaluation tool. The
BI is a calculated numerical index that allows comparing
performances of drilling bits in different operating scenarios,
e.g., drilling with reduced (or extreme) parameters, typical of
directional drilling technologies (Steerable Systems, Straighthole Drilling Devices, etc.), taking into account also the
evaluation of the dull bit.
Two types of indicators generally govern the process of
drilling, the qualitative and the quantitative variables.
Qualitative variables describe the state of the well bore and the
ability or potential to obtain good geologic or exploratory
data. Quantitative variables represent those parameters that

SPE 94798

can either be controlled or calculated during the drilling


operation. Some of the variables described above can be then
combined into a general Performance Equation (PE) of the
following form2:
PE = f (WOB,RPM,TQ,Q,H,BD,FP,DG,ROP,P, FM) (1)
Where WOB, RPM, TQ, Q, H, BD, FP, DG, ROP, P and
FM represent the weight on bit, revolution per minute, torque,
flow rate, footage, bit diameter, drilling fluid properties, bit
dull grading, penetration rate, pressure drop across bit nozzle
and formation properties, respectively. Obviously, the more
the variables considered, the more complicated the
relationship expressed by eq. (1). The variables defined in eq.
(1) are then separated into two groups: independent and
dependent variables. The independent variables have preset
values or can vary within the practical operating ranges once
the drilling process is defined. Dependent variables, on the
other hand, are determined by the application or by the
fluctuation of the independent variables. With reference to the
functional relationship expressed in eq. (1), WOB, RPM, P,
Q, BD, FP, FM are the independent variables, while ROP, TQ,
H, DG, are the dependent variables.
The proposed BI methodology considers a simplified
general PE in the following form:
PE = f (WOB, RPM, Q, HSI, ROP, TQ, H, DG)

(2)

Where WOB, RPM, Q, HSI (bit Horsepower per Squared


Inch) are the independent variables, and ROP, TQ, H, DG are
the dependent variables. Bit diameter is not considered as an
independent variable, and comparisons between bits of
different diameter have been performed by normalizing the
data set by unit diameter. In particular, values of WOB, Q, and
TQ per inch have been used. This is definitely a
simplification, and its influence on the precision of the method
is currently under investigation. However, with this
approximation it is possible to compare bits drilling in the
same formation, but with different diameter, both inside a
single well and amongst different wells. Bit Index evaluation
is achieved through the following steps:
1) Acquisition of the data set relatively to the bits under
investigation (WOB, RPM, Q, HSI, ROP, TQ, H, DG). One
bit must be labeled as the reference bit, or benchmark bit. By
definition, the BI of the benchmark bit is set equal to one.
Criteria for benchmark bit determination are discussed in the
case study reported at the end of the paper.
2) Definition of the dependent variables, described as a
function of all the independent variables. In this way, for each
dependent variable it is possible to write a set of equations,
whose number is equal to the number of independent
variables. Indicating with Xi and Yj the independent and
dependent variables, respectively, these functions can be
expressed as:
Yj = f (Xi)

(3)

where i = 1, 2,, m and j = 1, 2,, n; m and n represent


the number of independent and dependent variables,
respectively. The total number of equations calculated from
eq. (3) is mn.
3) Assignation of a parametric constant (Cj) to each of the
dependent variables, describing the particular drilling

SPE 94798

scenario. These constants are a sort of weight associated to the


importance of each single dependent variable, and are
established accordingly to the Operators needs. Obviously,
constants Cj must comply with the following constraint:
j Cj = 1

(4)

4) Definition of a linear relationship between the


dependent and independent variables, as an approximation of
eq. (3):
Yj = KXi

(5)

Where K is a constant, while subscripts i and j have the


same meaning as above. The calculation of the BI is valid if
considering all the relationships between dependent and
independent variables as linear functions. Non-linear
approximations have not been considered in this paper, but
this possibility deserves more investigations. The benchmark
bit and the bit under investigation are evaluated using eq. (4)
and (5), under the conditions of equal (or normalized) bit size
and formation type. Considering the benchmark bit, the
relationship between dependent and independent variables
(i.e., eq. 5) can be written as:
Yj,b = Kb,xXi,b

(6)

Where subscript b represents the benchmark bit and


subscript x refers to the independent variable under
investigation. Relationships similar to eq. 5 are subsequently
written for all dependent variables, making possible the
calculation of the proportionality constants Kb,x in different
drilling conditions. Similarly, considering the bit under
investigation, the relationship between dependent and
independent variables (i.e., eq. 5) can be rearranged as:
Yj,z = Kz,xXi,z

(7)

Where subscript z represents the bit under investigation


and subscript x refers to the independent variable under
investigation. All the proportionality constants Kz,x are
established as above.
5) Eq. (7) is then rewritten to establish the relationship
between the dependent variables of the bit under investigation
and the independent variables of the benchmark bit as follows:
Yz,b = Kz,xXi,b

(8)

The left side of eq. (8) expresses the bit under


investigation, while the right side the benchmark bit. Eq. (8)
describes the behavior of the dependent variable of the bits
under investigation, based on the independent variable of the
benchmark bit. Equations like the above must be written for
each dependent variable. The proportionality constant Kz,x is
then derived from eq. (7) and substituted in eq. (8), obtaining:
Yz,b = (Yj,z/Xi,z) Xi,b

(9)

Eq. (9) allows calculating the numerical values pertaining


to all the dependent variables of the bit under investigation,
relatively to all the independent variables, obtaining m values.
These values are then averaged by:
Yz, r =

Yz , b
m

(10)

Where Yz,r is the average value of the dependent variable


of the investigated bit relatively to the benchmark bit. This
procedure is then repeated for all the n dependent variables
under investigation.
6) Finally, recalling the meaning of the constants Cj
assigned at step (3) and the constraints of eq. 3, the Bit Index
is defined as follows:
BI = C j

Yz ,r
Y j, b

(11)

Where Yz,r and Yj,b are calculated by means of eq. (10) and
eq. (6). Yz,r and Yj,b represent the value of the dependent
variables of the bit under investigation and the value of the
same variable of the benchmark bit, respectively. The BI of
the bit under investigation is then compared with that of the
benchmark (equal one, by definition). In this way, it is
possible to compare bit performances relatively to the
benchmark bit. If the best performer bit (i.e., in term of
footage and ROP) is set as the benchmark, then the calculated
BIs of the investigated bits are decimals included in the
interval zero to one. The reverse holds if the worst performer
bit is set as benchmark (i.e., BIs of investigated bits are
numbers larger than one). Obviously, in this case each
absolute value of the BI changes, but not the overall ranking
with respect to the benchmark.
The bit performance evaluation with the BI method can be
considered innovative because it considers a large selection of
drilling parameters affecting bit behavior and bit
performances. It is worthy reminding again that BI is
calculated in the hypothesis of linearization of eq. (3).
Drilling Bit Catalog
The BI was developed and tested by mining data from a
particular classification of Italian geological formations, aimed
to the geotechnical/geomechanical formation characterization
by means of a relatively large number of significant drilling
parameters. This classification is called Formation Drillability
Catalog (FDC).
The FDC was implemented since 1999 at the Exploration
& Production Division of Eni Group (Eni E&P), Italy, and it is
continuously updated with data collected from new wells.
FDC is aimed to assemble in an accessible and simple format
the drilling data regarded as particularly useful for post-well
analysis and for a possible classification of the formation
drillability. FDC represents a collection of about 300.000
raw data (more than 8000 records, each one listing 34 different
quantitative parameters), collected in more than 80 different
formations drilled in Italy from mid-1990 up to now.
The FDC is composed by basic tables. A single table
contains data concerning a single well, and it is formed by a
series of lines and columns. A line is a string of information
concerning a particular drilled interval, called Minimum
Interval (MI). Columns report the sequence of MIs ordered
vs. depth. In general, a MI is a sub-set of drilling data similar
to a single bit-run record, but limited to a particular interval
drilled. In this way, each bit run is split up into several MIs,
that are established by the change of one or more of the
following parameters: lithostratigraphy and/or formation, bit
type, average penetration rate. In particular, the lithological

variations are estimated according to the typical geological


and technical criteria of the Operative Geology unit of Eni
E&P, which allow the identification of the various geological
units and lithological types. Changes in penetration rates are
considered significant for MI characterization when the
variation is more than 1 m/h within a drilled interval of more
than 10 m. The MI represents a very detailed description of
the drilled formations22, at least in term of formation
classification.
A basic table is composed by a set of records, and it
defines the so-called geomechanical clustering. Each record is
formed by 5 qualitative and 34 quantitative parameters,
collected from the master log. Qualitative information are: Eni
E&P well code and well name, spud and end date, formation
name, identification of technical problems and descriptive
remarks for a better comprehension of the operating context.
The most important quantitative parameters are the following:
top, bottom and thickness of the MIs, lithology (accordingly
to Eni E&P codes), hole inclination and azimuth, sonic log
transit times, formation integrity test, leak-off test, drilling
technique (turbo drilling, rotary, sliding, straight-hole drilling
device, etc.), bit type, diameter and IADC code, trip-in and
trip- out depth, footage, average ROP, mud type and density22.
In the first stage of the FDC analysis, MIs were grouped
according to the three congruity criteria. 1) MIs with the same
drilling characteristics (e.g., referred to the same formation,
which is likely to have a constant drillability). 2)
Contiguous MIs, in term of depth; c) MI referred to the same
run of the same bit (e.g., MIs drilled with two different bits
cannot be grouped). In this way, it has been obtained the socalled Homogeneous Geotechnical Interval (HGI) clustering,
which split up the bit runs into larger intervals compared to
those deriving from the MI classification (Tab. 1). With these
guidelines, HGI groups all the contiguous and congruous
MIs, obtaining a reduction of bit run subdivision consequent
to MIs criteria, making BI calculation easier. In most cases,
bit runs pertain to several HGI, albeit some runs are
completely drilled within a single HGI. Data derived from the
HGI clustering were further supplemented with other relevant
parameters normally recorded while drilling (such as WOB,
RPM, HSI, flow rate, torque, dull bit grading), which are
necessary to the calculation of the BI. The combination and
processing of all the above data constitutes a new database,
called Drill-Bit Catalog (DBC)23.
Case study
The case study here presented analyzes the bit performances
of three deep wells catalogued in the FDC, with the aim to
compare and evaluate bits through the BI methodology. These
wells, recently drilled in Southern Italy, are particularly
significant for the complexity of drilled formations, the use of
different drilling technologies, the typology and number of
bits used and the variability of borehole inclination, making
bit selection a crucial point of both well planning and drilling
operation. Table 2 reports the main features of these wells.
As far as the BI calculation is concerned, Table 3 reports
the number of bit runs drilled through some selected HGI. It is
important to recall that, in general, a complete bit run pertains
to several HGI, albeit some runs are completely drilled within
a single HGI. From Table 3, it is possible to draw the

SPE 94798

following observations. In well A the larger number of bits


was run in the Flysch Galestrino #3 and #4 formation (17-1/2
and 14-3/4 section). In the 12-1/4 section, many bits have
drilled through cherty limestone. In well B the larger number
of bits was run in the Irpine Units (12-1/4 section) and in the
Apula Platform (8-1/2 section). In well C it has been
considered only the 5-7/8 section, drilled through various
lithostratigraphic members of the Apula Platform with roller
cones and fixed cutters bits. In general, inside the Irpine Units
it is possible to make a performance comparisons between
well A and well B, whilst in the Apula Platform it is possible
compare bits between well B and well C.
Bit Index calculation was performed by means of a
standard calculation spreadsheet (Tab. 4). Constant Cj (i.e., the
values of the parametric constant assigned to each of the
dependent variables) have been chosen accordingly to the
perceived importance relatively to the particular drilling
operation scenario under evaluation. The following values
have been selected and associated to the dependent variables:
CROP = 0.5, CH = 0.3, CTQ = 0.1, CDG = 0.1. Obviously, the
sum of these values equals one. All calculations have been
performed at constant Cj values.
Under these conditions, BI was calculated by means of eq.
(11) for each bit drilling through the selected MIs or HGIs. It
is important to recall that the BI is a relative number, whose
value is referred to a particular benchmark bit. In this study,
the benchmark was chosen in such a way as to have all
calculated BI larger than the unit value of the benchmark BI
(i.e., the benchmark is the worst performer bit, in term of
footage and ROP). Drilling bits have been identified with their
standard IADC Code, regardless of the Manufacturer or of the
trade name of a particular bit type, to avoid commercialism.
Obviously, for internal purposes each analysis can be traced
back to the original bit type. This approach is significantly
limitative, especially in the interpretation phase here reported,
since today the IADC Codes assigned by major Manufacturers
badly reflects the true design features of most of the bits,
especially inside the roller cones family (i.e., for the same
IADC Code, bit design can vary significantly from
Manufacturer to Manufacturer).
The highest number of bit runs drilled through the selected
HGIs is in the Apula Platform #3 (31 partial bit runs, with
complete runs inside the same formation), in the Irpine Units
#3 (11 partial bit runs, with complete runs inside the same
formation) and in Cherty Limestone #2 (13 partial bit runs),
drilled with roller cones bits. Table 5 reports the results of BI
analysis. For each investigated formation, the table lists the
IADC Code of the bit with the highest BI, which can be
considered as the best performer. Concerning IU3, AP3 and
AP5, the highest BI is that of a fixed cutters bit; in this case, it
has also been reported the best performer amongst roller cones
bits only.
Best performer bits in the investigated formations are the
following: FL3, bit IADC 425M; FL4, bit IADC 425M; CL2,
bit IADC 515M; IU3, bit IADC 517X (roller cones), bit IADC
M221 (fixed cutters); AP3, bit IADC 517X (roller cones), bit
IADC M422 (fixed cutters); AP4, bit IADC M422; AP5, bit
IADC 517X (roller cones), bit IADC M422 (fixed cutters).
Figure 1 reports BI values of the best performer bits in well A,
listed by diameter and formation type. BI values (and thus bit

SPE 94798

performances) differ significantly inside the lithostratigraphic


units of the same formation (Irpine Units, IU), when
considering different bit diameters. In this case, BI
methodology can help the bit selection, pointing out the best
bit for each formation (or lithostratigraphic unit) and for each
bit diameter. This is a direct result of the BI methodology by
means of the formation classification implemented in the
DBC.
Conclusions
The objective of this research is the analysis, implementation
and application of a drilling model for bit performance
evaluation named Bit Index. The need for such an
interpretative tool arises from the intrinsic difficulties to
evaluate bit performances (and thus well planning) in case of
medium-to-severe formation variability and directional
drilling conditions. BI methodology allows to compare
performances of bits drilling with reduced (or extreme)
parameters, typical of directional drilling technologies, taking
into account also the evaluation of the dull bit.
The model keeps into account several dependent and
independent drilling variables, and can be easily adapted for
more variables to be considered, constructing a mathematical
tool able to help during well planning, drilling and post-well
analysis. The BI was developed and tested by mining data
from a particular classification of Italian geological
formations, aimed to the geotechnical/geomechanical
formation characterization by means of a large number of
significant drilling parameters, called Drill-Bit Catalog. This
classification turned out to be extremely functional to BI
methodology, since it discloses and organizes formation
details that are not easily detectable or usable from bit records
or master logs, allowing for targeted bit performance
evaluations. Results of a preliminary case study showed that
BI methodology could help the bit selection, pointing out the
best bit for each formation and for each bit diameter.
The proposed model, although in its early stage, proved to
be economic and reliable. The quality of bit performance
analysis obtained with this model seems also more effective
than the traditional quick look analysis, performed on bit
records, or on the pure evaluation of the cost-per-foot.
Acknowledgment
The authors would like to thank Eni, E&P Division for
permission to publish data. We wish also to thank all staff
engineers for their assistance during this preliminary study
period.
References
1. Cordy, L. et al.: Cumulative Rock Strenght as a Quantitative
Means of Evaluating Drill Bit Selection and Emerging PDC
Cutter Technology, paper SPE 77217 presented at the 2002 SPE
Asia Pacific Drilling Technology, Jakarta, Sept. 9-11.
2. Perrin, V.P. et al.: Drilling Index - A new Approach to Bit
Performance Evaluation, paper SPE 37595 presented at the 1997
SPE/IADC Drilling Conference, Amsterdam, March 6-7.
3. Warren, T.M. and Sinor, A.: Drag Bit Performance Modeling,
paper SPE 15618 presented at the 1986 SPE Annual Technical
Conference and Exhibition, New Orleans, Oct. 5-8.

4. Hightower, W.J.: Proper Selection of Drill Bits and Their Use,


paper SPE 794, presented at the 1964 Mechanical Engineering
Aspects of Drilling-Production Symp., Fort Worth, March 23-24.
5. Vitter, A.L. Jr.: Bit Performance, paper SPE 4646 presented at
the 48th Annual Fall Meeting of the Society of Petroleum
Engineers of AIME, Las Vegas, Sept. 30-Oct. 3, 1973.
6. Rabia, H. and Farrelly, M.: A New Approach to Drill Bit
Selection, paper SPE 15894 presented at the 1986 SPE
European Petroleum Conference, London, Oct. 20-22.
7. Farrelly, M. and Rabia, H.: Bit Performance and Selection: a
Novel Approach, paper SPE 16163 presented at the 1987
IADC/SPE Drilling Conference, New Orleans, March 15-18.
8. Fear, M. J et al.: An Expert System for Drill Bit Selection,
paper IADC/SPE 27470 presented at the 1994 IADC/SPE
Drilling Conference, Dallas, Feb. 15-18.
9. Galle, E.M. and Woods H.B.: Best Constant Weight and Rotary
Speed for Rotary Rock Bits, presented at the 1963 Spring
Meeting of the Pacific Coast District, API Division of
Production.
10. Bilgesu, H.I. et al.: An Unconventional Approach for Drill-Bit
Selection, paper SPE 68089 presented at the 2001 SPE Middle
East Oil Show, Bahrain, March 17-20.
11. Raynal, J.C. et al.: Organization of Field Tests and Evaluation
of Tricone Bit Performance Using Statistical Analysis and Sonic
Logs, Journ. of Petr. Tech., 23, 1971, 506-512.
12. Dumas, C.F. and Maidla, E.E.: PDC Bit Se1ection Method
Through the Analysis of Past Bit Performances, paper SPE
21073 presented at the 1990 SPE Latin American Petroleum
Engineering Conference, Rio de Janeiro, Oct. 14-16.
13. Rabia, H.: Specific Energy as a Criterion for Bit Selection,
Journ. of Petr. Tech., 37, 1985, 1225-1229.
14. Gstalder, S. and Raynal, J.: Measurement of Some Mechanical
Properties of Rock and Their Relationship to Rock Drillability,
Journ. of Petr. Tech., 18, 1966, 991-996.
15. Mensa-Wilmot, G. et al.: Formation Drillability - Definition,
Quantification and Contributions to Bit Performance, paper
SPE/IADC 57558 presented at the 1999 SPE/IADC Middle East
Drilling Technology Conference, Abu Dhabi, Nov. 8-10.
16. Uboldi, V. et al.: Rock Strength Measurements on Cuttings as
Input Data for Optimizing Drill Bit Selection, paper SPE 56441
presented at the 1999 SPE Annual Technical Conference and
Exhibition, Houston, Oct. 3-6.
17. OHare, J. and Aigbekaen, O.A.Jr: Design Index: a Systematic
Method of PDC Drill-Bit Selection, paper SPE/IADC 59112
presented at the 2000 IADC/SPE Drilling Conference, New
Orleans, Feb. 23-25.
18. Xu, H. et al.: An Approach for Modelling the Field Drilling
Data, paper IADC/SPE 36385 presented at the 1996 IADC/SPE
Asia Pacific Drilling Technology Conference, Kuala Lumpur,
Sept. 9-11.
19. Xu, H. et al.: A Practical Method for Modelling Bit
Performance Using Mud Logging Data, paper SPE/IADC 37583
presented at the 1997 SPE/IADC Drilling Conference,
Amsterdam, March 4-6.
20. Bilgesu, H.I. et al.: A New Approach for Drill Bit Selection,
paper SPE 65618 presented at the 2000 SPE Eastern Regional
Meeting, Morgantown, West Virginia, Oct. 17-19.
21. Falcao, L.L. et al.: PDC Bit Selection Through Cost Prediction
Estimates Using Crossplots and Sonic Log Data, paper
SPE/IADC 25733 presented at the 1993 SPE/IADC Drilling
Conference, Amsterdam, Feb. 23-25.
22. Da Dalt, G.: Basic Table for the Geotechnical Clustering of the
Formations, Eni E&P Div., Internal Report, 2002.
23. Magagni, M.: Technical - Economical Analysis and Evaluation
of Bit Performances, MS Thesis, University of Bologna, 2004.

SPE 94798

MI#

Bit Diam.

Formation

IADC

Top

Bottom

HGI#

Bit Diam. Formation

IADC

Top

Bottom

1
2
3
4

8-1/2"
8-1/2
8-1/2
8-1/2

FM #3
FM #1
FM #1
FM #3

517
517
517
517

3373
3403
3413
3419

3403
3413
3419
3426

8-1/2"

2+3

8-1/2

FM #3

517

3373

3403

FM #1

517

3403

3419

8-1/2

FM #1

517

3426

3468

8-1/2

FM #3

517

3419

3426

8-1/2

FM #1

517

3426

3468

Table 1. Example of Minimum Interval (MI) clustering inside a single well. The FDC classification splits up a bit run into five
different MIs. Here it is possible to group MI #2 and #3, since the formation (FM #1) is the same. The clustering defines the HGI.

Well name
(spud date)
A (2000)
B (2001)
C (2001)

Investigated Interval
Top
Bottom
360
3240
354
5367
3212
4520

Hole diameter

Hole profile

Formation

17-1/2, 14-3/4,12-1/4
16, 12-1/4, 8-1/2, 6
5-7/8

V
V, S, H
V, H

FL, CL, IU
IU, AP
AP

Table 2. Main features of the investigated wells. V, S, and H indicate Vertical, Slant and Horizontal well profile, respectively.
Formation acronyms are the following: Flysch Galestrino Formation (FL), Cherty Limestone (CL), Irpine Units (IU), Apula Formation
(AP). Some Formations may be split up into different lithostratigraphic units, indicated with a number (e.g., FL3, CL2, etc.).

Well
A
A
A
B
B
B
B
B
C
C
C

Hole profile
Vertical
Vertical
Vertical
Build up
Slant
Slant
Build up
Horizontal
Vertical
Build up
Horizontal

Diameter
17-1/2
14-3/4
12-1/4
16
12-1/4
8-1/2
8-1/2
6
5-7/8
5-7/8
5-7/8

FL3
3T
3T
1T

FL4
3T
2T

CL2

Formation
IU3

13T

AP3

AP4

AP5

2T
6T
3P
2T
10T + 4P
1T + 3P

1T

1T
3T

2P

6P

2T
2T
5P + 1T

1P

Table 3. Total number of bit runs drilled through the selected HGIs analyzed in the case study, classified by diameter, hole profile and
formation type. T and P indicate roller cones and fixed cutters bits, respectively.

Bit Index Calculation Sheet


Independent variable (m)
WOB
metric tonne
RPM
Rev/min.
HSI
HP/sq. In.
Q
liter/min.

Bench. (b)
26
136
60
3371

Invest. (z)
22
152
177
3777

Dependent variable (n)


ROP
m/h
TQ
kgm
H
m
DG

Bench. (b)
15
12
80
0.17

Invest. (z)
20
15
240
0.50

Cj
0.50
0.10
0.30
0.10

ROPz,WOB = 23.64

ROP of investigated bit under the condition of benchmark bit


ROPz,RPM = 17.89
ROPz,HSI = 6.78
ROPz,Q = 17.85

ROPz,r = 16.54

TQz,WOB = 0.08

TQ of investigated bit under the condition of benchmark bit


TQz,RPM = 0.06
TQz,HSI = 0.02
TQz,Q = 0.06

TQz,r = 0.06

Hz,WOB = 283.64

H of investigated bit under the condition of benchmark bit


Hz,RPM = 214.74
Hz,HSI = 81.36
Hz,Q = 214.20

Hz,r = 198.48

DGbz,WOB = 0.59

DGb of investigated bit under the condition of benchmark bit


DGbz,RPM = 0.45
DGbz,HSI = 0.17
DGbz,Q = 0.45

DGbz,r = 0.41

Bit Index of Investigated bit (z) = 1.61 - Bit Index of Benchmark Bit (b) = 1.00
Table 4. Simplified example of BI calculation sheet.

SPE 94798

Best

Formation

Performer Bit

FL3

FL4

CL2

IU3

AP3

Roller Cones

435M

425M

515M

517X

517X

M221

M422

Fixed Cutters

AP4

AP5
517X

M422

M422

Table 5. Result of Bit Index analysis performed on well A, B and C. For each investigated formation, the table reports the IADC Code
of the bit with the highest BI, which can be considered as the best performer. Concerning IU3, AP3 and AP5, the highest BI is that of a
fixed cutters bit; in this case, it has also been reported the best performer amongst roller cones bits only.

Well A
5

FL4

FL3

17-1/2"

14-3/4"

IADC 517X

IADC 435M

IADC 435M

IADC 435M

IADC 425M

BI values

12-1/4"

Bit Diameter

Figure 1. BI values of the best performer bits in well A, listed by diameter and formation (lithostratigraphic units FL3 and FL4). BIs
(and thus bit performances) differ significantly inside the lithostratigraphic units of the same formation (Irpine Units, IU). This is a
direct result of the BI analysis by means of formation classification implemented in the DBC.