You are on page 1of 16

Energy 34 (2009) 1097–1112

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Modeling the time and cost to drill an offshore well


Mark J. Kaiser*
Center for Energy Studies, Louisiana State University, Energy Coast & Environment Building, Nicholson Extension Drive, Baton Rouge, LA 70803, USA

a r t i c l e i n f o a b s t r a c t

Article history: The objective in drilling a hydrocarbon well is to make hole as quickly as possible subject to the tech-
Received 11 April 2007 nological, operational, quality, and safety constraints associated with the process. These objectives are
Received in revised form frequently conflicting and depend on factors that are subject to significant private and market uncer-
24 January 2009
tainty. There is no way to identify all of the relevant characteristics of drilling operations, but through use
Accepted 19 February 2009
of statistical analysis and empirical modeling, it is possible to develop relations that characterize and
Available online 22 May 2009
benchmark drilling performance under a suitable set of assumptions. The purpose of this paper is to
develop the conceptual framework to model the time and cost to drill an offshore well and to illustrate
Keywords:
Cost estimation the methodology on a test set of wells in the Gulf of Mexico. The physical characteristics of the wellbore
Benchmark performance and operational aspects of drilling, including variables such as the drilled interval, horizontal displace-
Drilling models ment, aspect ratio, number of casing strings, and mud weight, serve as the primary descriptive factors in
Functional relations the functional relations constructed.
Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction in, the final cost to drill and equip a well is about twice the rig
dayrate. A well that takes 30 days to drill with a $200,000/day
To find and produce oil and gas reservoirs, holes must be drilled jackup, for instance, would be expected to cost about $12 million to
into the Earth. A hole made by a drilling bit is called a well, and the complete.
primary objective of making a well is to produce underground Although the physics of drilling is the same everywhere in the
fluids such as oil and natural gas, and to study properties of geologic world, wells vary widely in complexity and type. To evaluate the
formations. Drilling operations are complex and labor intensive, differences that exist in drilling a well and to compare performance,
and although many activities continue to be automated, most jobs it is useful to establish general functional relations of drilling
are still performed manually, 24 h a day, seven days a week, in all time and cost. To understand performance, it is necessary to isolate
weather conditions. The work is strenuous and hard, and per- the factors of drilling and to quantify how these factors influence
formed in traditional 12 h shifts on a 2-week on/off rotation, and the operation. Historically, much of the work on assessing drilling
only extreme weather or mechanical failure will shut down performance has focused on the identification and elimination of
operations. non-productive time, such as freeing stuck pipe, fishing, repairing
Offshore drilling requires a floating or bottom-supported rig to equipment, and waiting on weather. Cost estimation and perfor-
conduct operations. Although offshore rigs and facilities are func- mance evaluation are not usually made outside a small subset of
tionally similar to land operations, the remote locations, offshore wellbores because of the nature of the processes involved and
environment, and peculiar logistical requirements mean that uncertain operating environment, the pervasive impact of tech-
offshore drilling costs will be higher than onshore drilling for nology differentials, and the many unobservable characteristics
similar depth wells. For example, in 2006, the average offshore well that influence operations.
drilled in the U.S. was about twice as deep and four times as costly Two methods are commonly used to benchmark drilling
as the average onshore well [1]. Large capital expenditures are performance. The first method is based on experimental design and
required to drill an offshore well, with rates to rent the drilling rig controlled field studies. Typically, one or more parameters of dril-
costing anywhere from $50,000 to $500,000 per day, depending on ling are varied to examine the impact of the variable on output
the rig type, water depth, market conditions and offshore basin. measures, such as the rate of penetration or cost per foot drilled. A
When the cost of labor, fuel, materials and equipment are factored significant amount of experimental work has been done to study
factor effects on drilling time and cost, and a large number of
technical studies has been performed by industry (e.g., see Ref. [2]
* Tel.: þ1 225 578 4554; fax: þ1 225 578 4541. and the references therein). The second method to study factor
E-mail address: mkaiser@lsu.edu effects is based on an aggregate assessment of drilling data. In this

0360-5442/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.energy.2009.02.017
1098 M.J. Kaiser / Energy 34 (2009) 1097–1112

method, drilling data is sampled across many different operators conflicting and themselves depend upon several factors (Fig. 1). The
and wells, and relationships that correlate drilling parameters are formation geology at the site and the location of the target reservoir
established through empirical modeling. Both approaches have is a primary factor. Geologic formations vary across the world, and
their advantages and disadvantages, advocates and proponents. indeed, within the same producing basin. Hard, abrasive, and
The purpose of this paper is to develop the analytic framework heterogeneous formations typically have low penetration rates,
to construct empirical models of drilling time and cost. We provide frequent drill string failures, and significant deviation from the
background material in x2–4 to orient the reader on the funda- planned trajectory. Deep reservoirs are usually characterized by
mentals of drilling – how wells are planned, the equipment used in low permeability, high temperature and pressure, complex fracture
the operation, the basic mechanics of making hole, and contract growth and stress regimes, and contaminants such as CO2 and
fundamentals – and to provide the context in which the drilling hydrogen sulfide, which increase the complexity of the well and
models are constructed. The background material is essential to requiring operators to deal with a number of issues concerning
provide motivation for the model and its development. In x5, safety and operational performance.
a description of the factors that influence drilling is presented, The drilling methods used to make hole depend upon the
followed in x6 by the development of the cost estimation meth- geologic formation and the technology applied, the amount of
odology and model formulation. In x7, drilling time and cost information known about the formation, the experience and pref-
regression models are constructed for a test set of offshore wells erences of the operator, available equipment, and the drilling
drilled in the Gulf of Mexico between 2002 and 2003. Model results contractor’s experience and execution [4]. Well characteristics are
are presented and statistical relationships derived to illustrate the specified by the drilling plan, the location of the target reservoir,
methodology. Brief conclusions complete the paper. In Appendix A, and the conditions encountered during drilling. Site characteristics
a glossary of common drilling problems is provided, and in such as water depth, operator experience in the region, and envi-
Appendix B, additional material on drilling contracts are presented. ronmental conditions influence the operator’s decision regarding
the selection of the contract and rig type, which in turn, influence
2. Background performance metrics. Exogenous events such as stuck pipe, adverse
weather, and mechanical failure cannot be predicted but can have
2.1. Purpose of drilling a significant impact on the time and cost of drilling.

Holes are drilled in the ground to search for oil and gas, to 2.3. Well construction process
acquire information about the geologic formation, and to develop
hydrocarbon reservoirs. A hole made by a drilling bit is called a well. The well construction process consists of four stages: design,
A company acquires a lease or contract area based on geological and planning, execution, and analysis. The design and planning phases
geophysical data and conceptual plays, and invests in additional represent the foundation of well construction, and is usually initi-
data and manpower to refine their knowledge of the region. If the ated through the preparation of a drilling proposal by geologists
results of analysis are encouraging, exploratory drilling may result. and reservoir engineers.2 The proposal provides the information by
A team of geologists, geophysicists, and engineers selects the well which the well will be designed and the drilling program prepared,
site and the drilling target based on magnetic, gravity, and seismic and includes project team selection; well design; health, safety, and
surveys. A well plan and a cost estimate are performed, typically environmental quality; tendering, contracting, and procurement;
from the drilling engineer, who has the responsibility for gathering finance and administration; operations planning; and logistics.
the technical experts, letting contract, and ensuring the success of The drilling engineer prepares the drilling prognosis, and all the
the operation. The well is then drilled. information that is required to safely and efficiently drill the well,
Substantial resources of money and time are invested to identify including the well location and water depth, the vertical depth and
the geological target and the prognosis is confirmed or refuted on total measured depth, the depth of the expected reservoir sands,
the basis of the results of the drilled, logged,1 and tested well. downhole reservoir pressures, expected hydrocarbons, the pres-
During drilling, evaluation is made from the cuttings and reservoir ence of hydrogen sulfide or CO2, evaluation needs (logging, side
fluids that appear at the surface and from the drilling and coring wall coring, drillstem tests), special drilling problems, final dispo-
indicators. A DrillStem Test in which the pressure response of the sition of the well, and future sidetracking.
reservoir is recorded during short production periods may be per- The well is then drilled according to the drilling plan, usually
formed to gather additional information on the reservoir. The under a dayrate contract, although turnkey contracts – where the
results from exploratory drilling are evaluated and may result in drilling contractor drills the well for a ‘‘lump-sum’’ (fixed) price –
either an appraisal well program or an abandonment of the pros- are also employed [5]. Since the drilling budget represents
pect. If hydrocarbons are detected, the company will typically a significant part of the capital expenditures for a field, usually
confirm and delineate the field through additional appraisal between 40 and 60% of total development cost, drilling operations
drilling and well testing; if the field is judged to be economic, the are carefully planned and closely watched, and operators maintain
company will develop and produce the reserves in accord with its meticulous and detailed records of each well drilled. In order to
particular risk-reward strategy [3]. better understand the drilling operations – what worked and what
didn’t, and why – a post-mortem analysis may be performed.
2.2. Drilling operations are complex and dynamic

The objective in drilling is to make hole as quickly as possible 2


Oil and gas companies are typically organized according to function, such as
subject to the technological, operational, quality, and safety drilling engineering, formation evaluation, reservoir engineering, etc. The geologic
group usually recommends the wildcat well location, while the reservoir group
constraints of the process. These objectives are frequently
recommends the development well locations. The drilling engineering group
makes the preliminary well designs and cost estimates, and if the decision is made
to drill the well, the drilling group then prepares a more detailed well design and
1
In wireline logging, a logging crew inserts a recording device into the bottom of writes the bid specification. The drilling engineer recommends the procedures to
the hole, and draws it upward, measuring the electrical, radioactive, and sonic drill the well, and in many cases, the original well plan will be modified as drilling
properties of the formation rocks and their fluids. progresses to maximize the efficiency of cutting operations.
M.J. Kaiser / Energy 34 (2009) 1097–1112 1099

Expected and actual


weather; wave, current, Supply and demand of drilling
eddies; frequency, duration rigs; marine equipment;
and occurrence of operator preferences; terms
storm alerts; extreme of the contract

Market
Environmental conditions
conditions
Well type; total depth; vertical
interval; horizontal displacement;
Formation type, well geometry; well complexity;
complexity, sequence, number of casing strings; maximum
type; special problems Well angle; formation pressure;
Geologic characteristics well temperature
conditions
Drilling
time and cost

Exogeneous
events
Problem wells; Formation
mechanical failure; evaluation
force majeure
Coring; logging;
Drilling testing time
characteristics Site
characteristics

Bit type, size. mud


weight; casing scheme; Water depth; distance
underbalanced mode to shore; region/country;
soil, wave, current
conditions

Fig. 1. The time and cost to drill a well is influenced by a number of variables.

2.4. Cost estimation The primary time-dependent costs include the time required to
drill a well, which is influenced by the well plan (e.g., interval
Cost estimation is performed specific to the drilling prognosis. depths, number of casing strings, formation evaluation require-
The usual procedure is to decompose costs into general categories ments), rate of penetration, and problems encountered. Rig costs
of site preparation, mobilization and rigging up, drilling, tripping and other services such as support vessels, aircraft, mud logging,
operations, formation evaluation and surveys, casing placement, and rental tools, are also time-sensitive. The costs for drilling fluids,
well completion, and contingencies. Typically, several categories bits, cement, logging and other consumables have a time-driven
are specified, and the drilling engineer itemizes the expected time component, but are mostly influenced by well depth and downhole
and cost per category [6–9]. Each cost component is identified and conditions. Some consumable items such as the wellhead and
categorized into minor cost elements, and the percentage contri- casing will be a fixed cost. Mobilization, demobilization, and
bution of the total cost is computed to identify the key cost drivers. preparation are fixed costs determined by the location of the site
To improve the range of the estimate, the uncertainty of the cost and the rig release location.
drivers is frequently quantified [10]. This forms the framework of Functional categories may also be employed to classify costs,
the well budget which is then sent to management for an Autho- with groupings that include pre-spud, casing and cementing,
rization for Expenditure (AFE) to drill the well. drilling rotating cost, drilling non-rotating cost, and trouble. The
In an AFE, intangible drilling and equipment costs, completion drilling rotating cost category includes all the costs incurred while
costs if the well is successful, and plugging and abandonment costs the drill bit is rotating such as bit and mud cost. Drilling non-
if the well is dry, are listed. The AFE typically includes estimates for rotating costs include tripping, well control, waiting, supervision,
the cost of the drilling rig, mud, logging, testing, cementing, casing, and maintenance. Drilling problems are grouped together as
well stimulation, prime movers, pumps, tubing, separator, and trouble and include stuck pipe, fish, lost circulation, hole stability,
other services and equipment required [11,12]. In joint operations, casing and cement problems. In Appendix A, a glossary of common
the operating agreement typically requires that the operator get drilling problems is provided.
approval from the non-operators for drilling expenditures. AFEs Specialized services such as perforation and cementing are
inform non-operators as to the drilling plans, providing cost esti- charged on the basis of a service contract, which involves both time
mates, and obtaining necessary approvals. and volume factors. Helicopter service will have an aircraft lease
rate per month, a flying charge per hour, and a fixed cost for
2.5. Cost components mobilization/demobilization, or may be contracted on an annual
basis. Logging operations will typically combine fixed costs for
There are a number of ways in which drilling cost can be mobilization/demobilization, time-dependent costs for tool rentals,
classified, based on functional category, time or depth dependence, and time-independent costs for tool charges.
or variable or fixed cost classification. Cost components usually fall The proportion of cost in each category will vary from well to
across more than one category and allocation schemes are well, but typically, the time-dependent proportion of total cost
company-specific. varies between 40 and 70% of the total cost. For offshore wells with
1100 M.J. Kaiser / Energy 34 (2009) 1097–1112

high rig and transportation costs, the proportion is toward the and are capable of drilling on a wide variety of tracks in water depth
upper range, while for onshore wells with relatively low rig day- up to 500 feet. For short distances, the platform is towed to site,
rates but high time-independent costs, the time-dependent vari- while for major moves, the rig is transported as cargo on a heavy lift
able cost will be toward the lower range. In the functional vessel. Once in position, the legs are lowered, hoisting the drilling
categorization, rotating drilling costs and casing/cementing are platform above the water. Jackup rigs are either ‘‘mat-supported,’’
usually the dominant costs overall increasing with drilling interval with the jacket legs attached to a submerged mat, or ‘‘independent-
and water depth. leg,’’ where the individual legs are driven down independently into
the ocean floor. Rig selection depends on availability and seafloor
3. Mobile offshore drilling units conditions. Independent-leg rigs are capable of working in deeper
water and harsher environments than mat-supported rigs. After
Offshore oil production began in 1947 when Kerr McGee Oil preloading the legs, the hull is raised above the sea surface to
Industries drilled the first producing well beyond the sight of land a height (called the air gap) that depends on the expected height of
in 18 feet of water, ten miles from the Louisiana coastline [13]. Some the waves or the height of the platform. A slot in the hull allows the
25 years later, offshore wells were being drilled in 1000 feet water wellhead to be positioned under the rig floor, or the rig floor and
depth, while today, drilling occurs in water depth over 10,000 feet support structure can be extended (cantilevered) from the side of the
deep [14]. Offshore drilling operations are significantly more hull to the desired drilling position. Independent-leg cantilevered
expensive, uncertain, and risky than onshore due to the harsh and rigs are usually priced at a premium over mat-supported slot rigs.
unpredictable operating environment. In water depths greater than 500 feet, semisubmersibles and
Oil and gas reservoirs are found in a wide variety of geographical drillships (also called floaters) are used. Floaters require specialized
and geological environments, and the rig which is used to drill technologies that are not used with bottom-support rigs, such as
a particular well must have the capability to drill in the environ- dynamic positioning systems, marine risers, and drill string motion
ment in which it is to be used to the depth required to reach the compensators. A semisubmersible rig (semi) is designed to float in
reservoir. Exploratory drilling is most often performed from the water and is held in position by multiple anchors or equipped
a mobile offshore drilling unit (MODU) such as a jackup, semi- with dynamically positioned thrusters (Fig. 3). Semisubmersibles are
submersible, or drillship. In the Gulf of Mexico, all of these types are very stable during high seas and winds and the most modern semis
employed. can drill in 10,000 feet of water. Semis are normally self-propelled
In shallow bays, marshes, and other protected waters where and supported by vertical columns on submerged pontoons. By
weather conditions are not harsh, drilling rigs are mounted on steel varying the amount of ballast water in the pontoons, the unit can be
barges that are pushed into place and flooded to sit on the bottom raised or lowered. The lower the pontoons lie beneath the surface,
of the water. These rigs are known as the ‘‘inland’’ rig fleet (or the less they are affected by wave and current action.
swamp barges) and typically work in 8–20 feet of water. A drillship has a conventional ship hull with a large aperture
Submersibles consist of a drilling platform mounted on vertical known as a ‘‘moon pool’’ through which drilling takes place (Fig. 4).
columns attached to pontoons. When water is pumped into the Drillships built before 1975 drilled in shallow waters while moored
lower hull, the rig submerges and rests on the seabed when drilling, in place, but went out of favor in the late 1980s. A new generation of
and when emptied, allow the rig to float and be tugged from one drillships built after 1975 were dynamically positioned which
location to another. Swamp barges operate in swamp and marsh allowed drilling in water depths up to 10,000 feet.
areas in Louisiana, Nigeria and Indonesia.
A jackup rig is a barge with legs that can be lowered or raised 4. Making hole
(Fig. 2). Jackup rigs are the most common MODU used worldwide
4.1. Stages of drilling

The start of drilling a well is called ‘‘spudding in’’. To spud in,


a large pipe, called the conductor or foundation casing, is either

Fig. 2. Mobile offshore drilling unit – jackup. Source: Pride. Fig. 3. Mobile offshore drilling unit – semisubmersible. Source: Transocean.
M.J. Kaiser / Energy 34 (2009) 1097–1112 1101

Fig. 4. Mobile offshore drilling unit – drillship. Source: Stena Drilling.

drilled, jetted, or hammered into the seabed from 100 to 500 feet to
a point below the drill floor. The conductor serves as the top part of
the well and creates a hole for the drillstem and casing to be low-
ered into.
Wells are drilled in stages:

 The bit and drill string is inserted into the hole and drills to
a certain depth.
 The drill string is removed from the hole.
 Casing is put into the hole to line it, and in most cases, is
cemented to the wall of the hole.
 The bit reenters the hole and the process is repeated until the
target is reached.

The drill string and bit represent the business end of drilling,
and all the equipment on the surface and subsurface is used to
support drilling with the bit. Wells are drilled ‘‘open hole’’ until it is
necessary to run and cement casing to protect the integrity of the
well. Wellbore stability can be a mechanical problem, where weak Fig. 5. Typical well configuration schematic.
formations exist, or it can be a chemically based problem, where
clays in the shales and other formations are weakened when 00 00
exposed to the drilling fluid. The amount of open hole limits how perhaps finishing with 738 or 958 diameter at target (Fig. 5). Surface
long an interval can be and how long it may be safe to expose the casing protects freshwater acquifers, anchors the blowout preven-
formations to the drilling fluids. tion equipment, and protects the hole from potentially hazardous
A well penetrates many different types of rock formations shallow formations. Intermediate casing protects holes from
(zones) until total depth is achieved, and as drilling continues abnormally pressured zones and poor formations such as salt and
deeper into the earth, the operating environment becomes more sloughing shale. Most casing is thin-wall, 30 feet sections of
hostile and drilling becomes more difficult: temperatures and seamless steel pipe and the space between the casing and the
pressures increase,3 formations become more abrasive and harder, borehole is cemented to support the casing and to prevent the flow
complex stress regimes and fracture growth develop, and the time of underground fluids to the surface and/or into freshwater zones.
and cost of drilling increases, often substantially. The number of casing strings needed to drill a well depends on the
A well is usually spudded with a 3600 , 3000 , or 2600 casing, and as formation and often varies from 3 to 8 strings. Troublesome
the hole deepens the casing becomes progressively narrower, formations such as high pressure zones, sloughing shale, and
shallow water flows require more intermediate casing. A narrow
pore pressure–fracture gradient will also require more casing
strings, and because of the time involved to trip out the hole and set
3
High temperature problems are most frequently associated with the instru- casing pipe, the cost of drilling increases with the number of casing
mentation used to measure and control the drilling direction and with logging
strings.
equipment. A negative temperature gradient runs from the waterline to seafloor,
which turns positive below the mudline and increases with the pressure approxi- The drill bit is attached to 30 feet joints of steel pipe screwed or
mately linearly with distance. The average temperature increase is 2  F/100 feet joined together as they go downhole. The drill pipe may be pre-
below the mudline, but it can vary between 0.5 and 5.0  F/100 feet depending on racked into double and triple joints depending on the capability of
location [11]. Fluid pressure depends on the density of the overlying water and the drilling rig. The pieces of formation which are cut away is called
averages 45 psi/100 feet In the GOM coastline, the normal pressure gradient is 46.
5 psi/100 feet A ‘‘typical’’ 10,000 feet well should therefore have a temperature of
cuttings, and during the drilling process, drilling fluid (i.e., mud) is
about 200  F and reservoir pressure 4500 psi. Deep shelf prospects may have circulated down the drill string, through the nozzles in the bit, and
temperatures as high as 450  F and formation pressures in excess of 30,000 psi. then back to the surface through the annulus between the drill
1102 M.J. Kaiser / Energy 34 (2009) 1097–1112

string and the borehole walls. At the surface, the drill cuttings, silt, meet the demands of the project and which are in the general
and sand are removed from the drilling fluid before it is returned vicinity of the well to be drilled. The terms of the contract can be for
downhole through the drill string. one or more wells over a short- or long-term basis on either
a ‘‘dayrate’’ or ‘‘turnkey’’ contract (see Appendix B for further
4.2. The role of mud details on contract terms). After the contractors submit their formal
bids, the drilling superintendent and his team select the rig that
Drilling fluids, also referred to as mud, play a number of they believe will drill the well in the most cost effective and safest
important functions in drilling: to control the pressures that exist in way. The operator selects the best bid according to price, avail-
the wellbore at different depths, to carry the cuttings out of the ability, reputation, past experience, and other factors. The contract
hole, to lubricate the drill string, and stabilize the wellbore. Drilling winner is not necessarily the lowest bid contractor.
mud consists of four basic parts: (1) base fluids – water, oil, Drilling contracts are complex instruments because in addition
synthetic material, or varying combinations – which classify the to covering the financial, legal, environmental and health and
mud; (2) active solids – the viscosity building part of the system, safety aspects of the job, they must also accommodate the uncer-
often bentonite clays; (3) inert solids – the density building part of tainties inherent in the operation, so that neither party is unduly
the system, such as barite; and (4) other additives – to control the penalized for unforeseeable events [16]. Offshore contracts origi-
chemical, physical, and biological properties of the mud, such as nated from their land-based counterparts and developed over the
polymers, starches, and various other chemicals. years with increased sophistication, but there is still little stan-
Mud is classified into three general categories: water-based mud dardization. The American Petroleum Institute, the International
(WBM), oil-based mud (OBM), and synthetic-based mud (SBM) Association of Drilling Contractors, the American Association of
[15]. WBM is made with fresh or saline water and is used for most Drilling Contractors, the United Kingdom Cost Reduction Network,
types of drilling, often consisting of dissolved salts, additives, and other organizations all have their own draft drilling contract
polymers, clays, and weight material such as barite. OBMs are forms. Operators also maintain forms [17] which are designed to
water-in-oil emulsions with dispersed clay and weighting material. avoid problems encountered in their particular experience.
In SBM, the oil is substituted with long-chain esters, ethers, acetyls, A drilling contract is a service contract in which a company
and synthetic hydrocarbons. OBM and SBM have special opera- agrees to perform certain services for a monetary payment. The
tional advantages over WBM due to their low friction, good payment terms under a dayrate contract are usually decomposed in
temperature tolerance, and inertness to formation clays. SBM were terms of four basic subrates: mobilization rate, operating rate,
developed in response to the requirements for drilling fluids with reduced rate, and special rate.
performance comparable to OBM but without the adverse envi- The mobilization rate is used during the time the rig is mobi-
ronmental implications. SBM and OBM are not generally used for lized/demobilized between well locations or shore, and covers port
the entire depth of the well; typically, WBM are used for the upper fees, towing cost, fuel and other expenses incurred by the
portion with a change to SBM below the 1600 or 1300 casing point at contractor to arrive at site. Mobilization/demobilization may also
a depth of 5000 feet or more. Each mud program is well specific. be specified on a lump sum basis.
Formation pressure can be unpredictable, and therefore, The operating rate governs the cost during drilling and covers
potentially hazardous. If the drill bit penetrates a high pressure the rig rental, crew and consumables. The dayrate is primarily
zone unexpectedly, oil or gas, or a mixture of both, may rush into determined by water depth and rig capabilities, the supply/demand
and up the wellbore, dilute the mud, and reduce its pressure. This is conditions that exist in the region at the time the contract is
called a ‘‘kick,’’ and it can lead – if unchecked – to an uncontrollable written, and the duration of the contract. Most shallow water rigs
gusher at the wellhead (a blowout). Drilling fluid and the experi- are relatively non-specific assets, capable of drilling on a wide
ence of the drill team is the first line of defense to prevent this variety of tracts. Deepwater rigs are more capital intensive and
occurrence. The last line of defense situated below the wellhead sometimes firm-specific.4 The drilling business is highly competi-
control valves is the emergency blowout preventer (BOP); in tive and dayrates typically follow utilization levels.
deepwater, BOPs may be situated on the seabed. Kicks and blow- The reduced or standby rate covers conditions when the rig is
outs can be detected by monitoring the density, viscosity, and other not drilling, such as during moves, logging or testing, while
properties of the drilling mud. A drilling problem may require special rates are used when drilling is interrupted by adverse sea
chemical additives (‘‘pill’’) to be injected into the mud, or heavy or weather conditions, mechanical failure, force majeure, or the
mud (‘‘kill mud’’) to circulate the kick out of the well. concession holder’s inability to obtain the required permits or
The density of the mud depends on the formation pressures authorization.
anticipated or encountered, which in turn impacts the rate of bit The total drilling cost under a standard dayrate contract is
penetration. Heavy mud retards bit penetration and can cause stuck determined thus
pipe and skin damage. Light muds are used if the well is planned
4
X
and drilled in underbalanced mode, where the pressure of the fluid
column is designed to be less than that of the formation pressure. DHC ¼ Ti Ri ;
i¼1
Drilling is usually performed underbalanced in non-permeable
zones and overbalanced in permeable zones. In development dril- where DHC ¼ total drilling cost ($), T1 ¼ total amount of time rig
ling, the local pressure regime is usually known, while in wildcats, mobilized and anchored (h), T2 ¼ total amount of drilling time
it is uncertain since the local geology has not yet been explored. (day), T3 ¼ total amount of time spent moving, logging, or testing
(h), and T4 ¼ total amount of downtime spent on equipment failure
4.3. Generalized dayrate contracts or waiting on weather (h), R1 ¼ mobilization rate ($/h),
R2 ¼ operating rate ($/day), R3 ¼ reduced rate ($/h), and R4 ¼ special
The operator writes a drilling prognosis – which essentially is rate ($/h).
the recipe book for the well, and includes equipment and proce-
dures that the operator will require, together with a well descrip-
tion, bid specifications, and drilling contract – and then the drilling 4
Some E&P companies participate in joint ventures with drilling contractors to
superintendent will choose a group of companies with rigs that finance, design, build, and operate deepwater rigs.
M.J. Kaiser / Energy 34 (2009) 1097–1112 1103

Physical characteristics
Measurable of the well, geology, and
(Direct) drilling process

Observable

Measurable Operator experience,


(Indirect) well quality, operator
characteristics

Drilling
Factors

Well planning and execution,


Unobservable comunication, team
dynamics, leadership,
project management

Fig. 6. Drilling factors are classified into observable and unobservable categories.

Contract terms are proprietary, but dayrates are widely reported Exploratory wells are drilled to find oil and gas, and their
and tracked by industry. The time for the rig to mobilize and anchor primary purpose is to gather information on subterranean condi-
depends on the type of rig and the distance from its current loca- tions and confirm whether geological formations contain hydro-
tion to the new site. The total amount of time spent coring, logging, carbons. Most exploratory wells are drilled as straight as possible,5
and testing depends upon the well type, well design, and the but in special circumstances, may be drilled at an angle or even
success of drilling. The amount of time spent moving the rig or horizontally. Development wells are drilled as part of a production
waiting on weather is a stochastic function depending on the region plan. Usually only the first development well is vertical; subsequent
and time of the operation. Downtime spent on equipment failure or wells are drilled vertical to a certain depth and then kicked off in
problem wells may be attributable to geologic conditions, operator a ‘‘J-shaped’’ or ‘‘S-shaped’’ pattern to total depth. A planned
and contractor experience, and a random forcing function. multilateral will involve sidetracking out of the well from a zone
shallower than the original leg of the wellbore to reach a new
5. Model factors section of the formation.
The first exploration well in an area will be drilled very carefully
Drilling operations are complex and uncertain, and many factors because the geologic formations are untested and the risk of
influence the time and cost to drill a well. These factors are overpressure may result in a blowout, but after a few wells the
multidimensional, often interdependent, and usually stochastic; stratigraphic layers where overpressures can be expected are
the factors can be classified as either observable or unobservable known, and drilling can often proceed at a faster rate. The time and
(Fig. 6). Measurable factors include the physical characteristics of costs to drill a development well is expected to be smaller than
the well, geology, and drill parameters; while indirect character- exploration wells because information gathered from exploration is
istics, such as wellbore design, contractor experience and hole applied in drilling. Learning economies are also well documented in
development drilling [18].
quality, need to be proxied through other variables if they are to be
incorporated in analysis. Factors such as communication, leader- Well status describes if the well hit ‘‘pay dirt’’ (successful) or was
a ‘‘duster’’ (dry hole). An unsuccessful well will often lead to side-
ship, and project management skills will also impact drilling
performance, sometimes significantly, but to capture and identify tracking to test a different section or zone. Dry wells will be plugged
with cement and abandoned, and the extra time and cost of this
the influence of these (unobservable) variables is usually beyond
the scope of analysis and the reporting capabilities of operators. The operation suggest that dry hole cost will exceed, if all other things
are equal, the cost to drill a successful well. This is generally true for
amount of data required to construct a reasonable representation of
drilling operations depends in part on the nature of the data set and onshore wells as reported by American Petroleum Institute drilling
cost data [1,19], but in an offshore environment, the extra cost to
the system characteristics.
plug and abandon a well is usually a small part of the total drilling
cost and is not easily distinguished. Offshore, wildcats that find
5.1. Well characteristics

5.1.1. Well type


5
There are many ways to classify wells. The most common Because the bit is being turned clockwise during drilling, the wellbore tends to
distinction is between exploratory and development wells. Wells ‘‘walk to the right’’ with a clockwise corkscrew pattern as it goes down. Drilling
contracts often have a clause stipulating that the well can change in deviation no
that are drilled in an unproved area to add reserves are exploratory more than 3 per 100 feet anywhere along the wellbore and must end up in a cone
wells, while wells drilled in the known extent of a field to produce with a maximum angle of 5 . This is commonly referred to as a ‘‘straight’’ hole, but
known reserves are development wells. other definitions exist [11].
1104 M.J. Kaiser / Energy 34 (2009) 1097–1112

a commercial pay zone may be temporarily plugged and aban- between 2 and 3 times the cost of a similar length vertical borehole.
doned until the field is developed or permanently plugged if the Directional wells with long horizontal departure are called
location is not optimal from which to produce the field. extended reach wells and are often defined as wells with TD/
VD > 2.0.
5.1.2. Well geometry
A wellbore is a three-dimensional object which can be described 5.1.3. Casing geometry
in geometric terms with respect to the length, diameter, and Casing serves several important functions in drilling and
curvature of the hole trajectory. The depth of a well measured from completing a well, and is one of the most expensive parts of
the rotary table in the XY-plane along the length of the wellbore is a drilling program, ranging from 10 to 20% of the total cost of
called total depth (or total measured depth) and denoted by TD, a completed well [20]. Casing prevents collapse of the borehole
while the (true) vertical depth VD is the distance from the rotary during drilling and isolates the wellbore fluids from the subsurface
table measured in a vertical plane to TD. Spud depth SD is the formations. Casing also provides a flow conduit for the drilling fluid
distance from the rotary table to the seabed. The drilled interval DI and permits the safe control of formation pressure. A well that does
is the difference between the total depth and the spud depth, not encounter abnormal formation pore pressure gradients, lost
DI ¼ TD  SD; the vertical interval is defined as VI ¼ VD  SD. circulation zones, or salt sections usually require only conductor
The horizontal displacement HD is the distance in plan view6 from and surface casing to drill to target. Deeper wells that penetrate
the rotary table to TD. Water depth WD is the distance from the abnormally pressured formations, lost circulation zones, unstable
waterline to the seabed. shale sections, or salt sections generally will require one or more
The problems, costs, and hazards of drilling increase with water strings of intermediate casing to protect formations and to prevent
depth and drilled interval. Water depth is an important factor in all problems [21,22].
offshore operations because as water depth increases, rig specifi- Each casing section has a diameter Di ¼ D(Si), radius Ri ¼ R(Si),
cations also need to increase. Drilling operations with floaters and length Li ¼ L(Si) measured from the rotary table. If k ¼ NS
require even more specialized technology which adds significantly represents the number of strings associated with the well, then
to dayrates. The deeper the hole the more time is lost in round trips a finished well is characterized by the vectors D ¼ (D1, D2,., Dk) and
to replace worn bits and to run casing, tests, and logs. The number L ¼ (L1, L2,., Lk). The number of casing strings provides an indirect
of formations encountered will typically increase along with the measure of well complexity, since complex wells are frequently
number of casing strings required to maintain well control. As the associated with multiple strings and narrow margins between pore
number of casings increase, the trip time, installation, cement, and pressure and fracture gradients. The incremental well casing length
cementing time will also increase. is denoted by L* ¼ ðL*1 ; L*2 ; .; L*k Þ, where L*i ¼ Li  Li1 , for i ¼ 1,.,
The direct cost of the additional strings has an effect, but there k, and L*0 ¼ 0. If hole sections can be drilled without setting
are also costs that occur because of well-diameter constraints. The intermediate strings or liners,7 then drilling can proceed quickly.
ability to handle larger casing requires more expensive rigs, tools, Operators generally prefer the production casing to be as large
pumps, compressors, and wellhead control equipment. Increasing as possible to maximize production, but large production casing
the number of casing strings from 3 to 4 may result in a 10–20% requires a large wellbore, which is more complicated and expensive
increase in well cost; increasing the number of strings from 4 to 5 to drill, since the rig has to be higher spec and more rock volume
may increase cost by 20–30%; and so on. Offshore drilling mud is an has to be removed. Average hole size and removed rock volume are
expensive chemical preparation, and the deeper the hole, the more numerical measures that characterize the geometry of the final
mud is needed. Beyond a certain depth below mudline drilled well. The average hole size HS is determined by the
(15,000 feet), technical complications and the opportunity for weighted average diameter of the casing string along the wellbore:
problems increase significantly. In many deepwater wells, the
Pk
percentage of total well cost as a fraction of total depth is such that Di L*i
as much as 50% of the total cost can be spent on drilling the last 10–
HS ¼ Pi ¼k 0 ;
*
i ¼ 0 Li
20% of the well.
A well is composed of segments of casing string Si oriented at an while the rock volume removed VR from the wellbore without
angle A(Si) relative to a reference coordinate system. The maximum washout is defined as
angle of the wellbore is computed as MA ¼ max AðSi Þ. If MA  85
i k
the well is classified as a horizontal well using the indicator variable X
HW: VR ¼ p R2i L*i :
i¼0
 
1; MA  85 The cost of drilling should be larger the greater the average hole
HW ¼
0; otherwise: size and volume removed from the wellbore.
If L(Si) denotes the length of well section Si, then the total length
of the horizontal section of a well is denoted as HL: 5.1.4. Well complexity
A wide variety of well types and configurations exist, along with
( ) several notions of what constitutes a ‘‘complex’’ well, and it is
l
X  
HL ¼ LðSi ÞAðSi Þ  85 ; i ¼ 1; .; l : unlikely that a single definition will ever be widely accepted since
i¼1 practices, opinions, and experiences among drilling contractors
Oil and gas wells are drilled horizontally for a variety of reasons,
but primarily to improve production and reach reserves that
otherwise might not be developed. Horizontal drilling is less stable 7
Liners are casing string that are suspended from the bottom of the next largest
than drilling vertically, more difficult to log and complete, and often casing string. The principal advantage of a liner is its lower cost, since it does not
run the length of the well, but problems may arise in production and workovers
from hanger seals and cement leakage. The decision to use strings or liners
represents a trade-off between the cost savings realized by using less casing versus
6
Looking down from above as if the well path is straight. potential future problems caused by hangers and leakage.
M.J. Kaiser / Energy 34 (2009) 1097–1112 1105

vary significantly. Complex wells arise from geologic formations, (dayrate, turnkey), job specification (one well, multiple wells),
target depth and the size of the reservoir sands, wellbore trajectory, market rate, and negotiating strategies are important factors in
the experience of the contractor and application of technology, as determining drilling time and cost.
well as numerous other factors. Conditions that create a complex
well are often proxied through a description of the physical char- 5.3.2. Rig selection
acteristics of the well, such as the water depth, vertical depth, Many different rigs can be used to drill an offshore well. Rig
maximum angle, and number of casing strings. selection depends upon factors such as the type of well being drilled,
High-pressure/high temperature HP/HT wells begin to exhibit water depth and environmental criteria, the type and density of
high temperatures at depths around 18,000–20,000 feet, although the seabed, expected drilling depth, load capacity, frequency of
in areas with geothermal activity, hot drilling conditions can occur moves, ability to operate without support and availability.
at shallower depths. HP/HT wells are usually planned and drilled Before selecting a rig, a detailed site assessment is completed to
using significantly less formation data than shallower and cooler identify water depth and bottom conditions, and expected weather,
wells. The trajectory of a well, G, can also be used to describe wind, tides, wave heights and current speeds. Water depth is an
complex wells. A well is often considered complex if its formation important factor since the rig must be rated to work in the depth
pressure FP exceeds 10,000 psi or temperature T is greater than required; e.g., a jackup must be able to jack up on location and
300  F anywhere along the wellbore. A complexity index is used to withstand the normal environmental forces encountered. Seabed
identify complex drilling environments if any of these conditions conditions determine if an independent-leg or mat-supported
occur: jackup can be employed. Independent-leg jackups are employed in
 firm soil, coral, or uneven seabeds; mat-supported jackups require
1; FP > 10; 000 psi; T > 300  F; or G complex low soil shear values and a flat seabed. The decision to use an
CI ¼
0; otherwise: anchored or dynamically positioned rig depends on the nature of
the seafloor, the size of the rig, and the expected operating condi-
The ratio of the horizontal length to the total footage drilled
tions. Deepwaters characterized by strong currents create a need
describes the percentage of the well’s footage drilled under hori-
for high specification rigs capable of maintaining station, and in
zontal conditions:
some instances, suppressing vortex induced vibrations.
HL If weather and environmental conditions are expected to be
HR ¼ ; a problem, then sophisticated all-weather semis can be used to
DI
hedge against weather downtime. The increase in availability is
e.g., for most exploratory wells HRz0, while for development and achieved through the higher capital cost of the equipment; which in
extended reach wells, 0 < HR  1. The aspect ratio AR, turn is passed to the operator in higher dayrates. Jackups are
cheaper but are more prone to weather delay. The choice is up to the
HD
AR ¼ ; operator: the trade-off is between drilling availability and dayrate.
VI
Rig availability also plays a role in the selection process, since if
measures the aggregate curvature of the well trajectory, and the the regional demand exceeds supply, market rates will rise. In the
extended reach ratio, ER, is defined as the ratio of total depth to past, dayrates were based on investment and depreciation sched-
total vertical depth: ules, but when a market has a capacity shortage, dayrates will be
determined primarily by demand rather than cost recovery. If
TD regional utilization rates are low, drillers may stack jackups and
ER ¼ :
VD contract large rigs to shallow water projects to maintain the utili-
All three ratios provide metrics that quantify the wellbore zation of high cost equipment.
trajectory.
5.4. Drilling characteristics

5.2. Site characteristics


5.4.1. Bit size and type
Different types and sizes of bits are used according to the
The well site is characterized by its geographic location deter-
hardness of formations, pressure regime, and drilling plan. Bits are
mined through its latitude/longitude coordinates, the distance
classified as roller bits, such as steel tooth and insert bits, and fixed
from the well to the nearest onshore service station, and the water
cutter bits, such as polycrystalline diamond compact (PDC), ther-
depth. Water depth and environmental conditions expected to be
mally stable polycrystalline (TSP) and natural diamond bits. For
encountered are primary determinants in the selection of the rig
hard and abrasive drilling conditions in deep wells drilled with
required for drilling. As water depth increases, larger and more
mud, tricone bits are commonly used, but their susceptibility to
robust rigs are required, with extra hoisting capacity, mud circu-
wear and bearing failure limits their drilling time. In deep hole
lation systems, mooring systems, etc. The region and country in
sections, where tripping times are longer, polycrystalline or natural
which a well is located is an important consideration in obtaining
diamond bits are competitive with the tricone bit [23]. The final bit
government regulations and permits. The maturity of the infra-
size employed in the well is denoted FBS.
structure support services can play a significant role in determining
drilling cost, and the knowledge and experience of the contractor,
5.4.2. Drilling fluid
say as measured by the number of wells drilled in the region, can be
Drilling mud is applied to control the pressures that exist in the
important factors.
wellbore at different depths. Synthetic-based muds have downhole
performance similar to OBMs and offer advantages over WBMs in
5.3. Operator preference certain circumstances; e.g., SBM have higher lubricity compared to
WBM and can endure more hostile downhole conditions. The use of
5.3.1. Contract type SBM is likely to enjoy faster penetration rates and is also less likely
The operator decides not only where to drill, but also how to than WBM to interact with the production horizons because
drill, and the manner in which to let the contract. The contract type their physical and chemical makeup is similar to the hydrocarbon
1106 M.J. Kaiser / Energy 34 (2009) 1097–1112

bearing zone. Other fluid-related issues involved in the choice of weather WOW time needs to be considered separate from the
drilling fluid system include hole cleaning, lubricity, stability, barite drilling time and cost metrics.
sag, and fluid stability. In horizontal, multilateral and extended
reach applications, fluid-related issues become more complicated, 5.6.2. Interrupt time
and the more complex the mud program, the greater the expected Many problems can occur during drilling requiring suspension
cost of drilling. of the activity. Most contracts specify a certain amount of ‘‘free’’
downtime (24 h per month is typical), but outside this allowance,
5.4.3. Mud weight the contractor does not receive payment for the time the rig was
Heavy mud is typically used to create an overbalance to prevent inactive. Other delays that may occur not directly accountable to
fluids from entering the well. The greater the hole pressure, the the driller are usually charged at a reduced rate.
heavier the mud, and the slower the drilling. Mud weights vary One of the most common problems in drilling a hole is that
over each well section. If the mud weight applied to drill out well something breaks inside the well, such as a piece of bit or drill
section Li*, i ¼ 1,., k, is denoted MW(Li*), then the average mud string, or something falls down into the wellbore, such as wrenches
weight employed in the drilling program is defined by or other tools (see Appendix A). These pieces of metal are called
  ‘‘fish’’ or ‘‘junk,’’ and because the bit cannot drill through them,
Pk
i¼0MW L*i L*i drilling must be suspended until they can be retrieved by ‘‘fishing’’
MWA ¼ Pk : tools leased from a service company. The use of fishing tools cost
*
i ¼ 0 Li the operation extra money and time spent fishing is flat time. If
Mud weight at TD is denoted MWTD. The maximum mud weight fishing is unsuccessful, the hole will either be sidetracked or
used over the wellbore is determined as whipstocked (drilled around the obstacle), and in the worst case,
a new hole may need to be spudded. Rig equipment failure, lost
 
circulation and stuck pipe are the main causes of downtime in the
MWMX ¼ max MW L*i :
i ¼ 1;.;k Gulf of Mexico, and in some cases have been reported to account for
nearly 25% of drilling budgets [26]. Interrupt time cannot be pre-
To the extent that high pressure wells are more complex to drill,
dicted with any degree of accuracy, and so an upper limit exists on
MWA, MWTD, and MWMX serve to proxy well complexity [24].
the expected reliability of any cost prediction tool.

5.5. Formation evaluation 5.6.3. Oil and gas prices


There is a correlation between hydrocarbon prices and drilling
Formation evaluation is a critical step in exploration since it is costs. When the price of crude oil increases, there is generally an
the stage in which information about the presence/absence of increase in exploration and drilling activity, and a decrease in rig
hydrocarbon bearing reservoirs is acquired. It is important to availability. Supply-and-demand lead to an increase in the costs of
remember, however, that time spent coring, logging, reaming, and rig rental rates, material costs, and services. Natural gas prices
testing is ‘‘flat’’ time, and so for wells that require extensive impact rigs drilling for gas on a regional basis.
formation evaluation data, normalization is required prior to
performance comparison. 5.7. Unobservable variables
Cutting a core requires separate round trips to install and
remove the coring assembly, and because coring is slower than Many variables which influence drilling performance, such as
drilling, if the well is deep and the interval is large, coring can incur well planning and preparation, project management, and technology
a significant amount of rig time. The number of days spent drilling are difficult (if not impossible) to quantify, and since the manner in
out core samples, or trying to core, the well in any hole section is which these variables impact the drilling program are essentially
denoted by CD. The number of days spent logging, or trying to log, unobservable, it is usually not possible to directly incorporate these
before or after total depth is reached, but excluding logging while factors into modeling. The importance of these factors to perfor-
drilling or formation evaluation drilling, is denoted LD. Coring and mance cannot be overstated, however, and is commonly recognized
logging days include interrupt time and waiting on weather time. as playing an important role in operations [27].

5.6. Exogenous events 5.7.1. Well planning


The first step in planning any well is to design the wellbore path
Exogenous events cannot be predicted with any degree of to intersect a given target. Were adequate time and resources given
accuracy, but their impact and duration needs to be considered to to the operational team to develop an efficient and best practice
normalize for conditions beyond the control of the contractor. well design, or was too little given too late for the team to succeed?
Careful planning and evaluation are required to successfully
5.6.1. Weather complete a project whether drilling easy, normal pressure wells, in
Offshore drilling may be subject to significant delays caused by shallow water, shallow target, benign environments; or difficult
the weather, and weather downtime can play an important factor in wells with a complex geometry that have a combination of high
the total costs of the operation. Weather downtime can impact temperature, high pressure, narrow pressure/formation gradient
drilling operations in various ways; e.g., weather too severe for windows, salt, rafted shale, high angle, shallow hazards, deepwater,
operations involving supply boats may lead to delay if stock levels contaminated environments. A multi-disciplinary operational team
on the rig decline to a critical level; weather may impact anchoring is usually the most efficient to deal with drilling/well construction
up and moving time; weather may be too severe for drilling to issues and objectives [28–30].
occur; and extreme weather may result in damaged or lost drill
strings and risers. If operating limits are exceeded because wave 5.7.2. Project management and leadership
heights, ocean currents, or eddies are too strong, drilling operations Comprehensive, forward looking, integrated engineering plan-
will be temporarily abandoned and resumed when conditions fall ning, coordination, execution and management, including defined
within the operating capabilities of the equipment [25]. Waiting on contingencies and options, helps the drilling program to be
M.J. Kaiser / Energy 34 (2009) 1097–1112 1107

executed in the shortest possible time [31–34]. What was the and test the model coefficients ða00 ; a01 ; .; a0p Þ for statistical
experience of the project management team and how was project significance.
management executed? Was leadership shown in decision making
and planning goals? Step 4. Maintain the significant variables and other factors of
interest, re-estimate the model specification,
5.7.3. Well quality
The timely delivery of logged, tested, and producible boreholes
q<p
X
is important to the profitability of the field [35–37]. What was the
quality of the final wellbore? Was there skin damage?
4ðUÞ ¼ a0 þ ai Xi
i¼1

5.7.4. Technology and report the final model results.


The impact of technology on drilling performance is pervasive Linear models are selected for ease in presentation and fix the
but difficult to isolate. Technology may be ‘‘enabling’’ or ‘‘enhancing’’ functional form. If linear models do not provide adequate response,
or both, and will normally shift from enabling to enhancing over more complex nonlinear models can be adopted. Because a number
time. New technology is expensive, both in terms of cost and of (unobservable) variables were omitted that may affect the time
learning, but if the technology reduces drilling time or improves the and cost of drilling, such as management efforts, well quality,
efficiency or safety of the operation and becomes widely adopted, technology and weather conditions, it is possible that omitting
costs decline and the performance efficiencies will improve and these variables may result in biased estimates if they are correlated
become absorbed within the process. Trade-offs between competing with other independent variables.9 We could use a fixed effects
technologies and processes are common,8 but quantifying the model to address the omitted variable concern by including loca-
differences in impact remains notoriously difficult to evaluate. tion, time, and developer fixed effects, and if the sample is large and
sufficiently robust, this would be reasonable to investigate.
6. Model formulation
6.2. Dependent variables
Associated with every well w is a vector of dependent variables,
4(w), and a matrix of observations of independent variables, X(w). The dependent variables of the model include the time to drill
The dependent variable set includes both reported and derived a well described by the number of dry hole days, DHD, the number
measures. The matrix of observations is used to explain the of days from the spud date to TD, plus any time spent batch setting,
dependent variables and is selected based on the underlying pre-installing or pre-setting casings. Dry hole days are normally
engineering and mechanical processes associated with drilling, the defined to include interrupt and weather time, time spent on
choice of dependent variable, data availability, level of aggregation, technical/mechanical sidetracking, and the operational time to set
and other factors. a whipstock in multilateral wells. The total number of days from the
rig arrival on location until the rig is released is the total site days,
6.1. Methodology TWD. This time includes the time for mooring and demooring,
where applicable, completion and testing, suspension and plug and
A large number of parameters influence drilling operations and abandonment activities.
interact in a complex and dynamic manner. Regression analysis The cost of operations during the dry hole day period plus the
provides a standard and transparent analytic framework to estab- casing batch set, pre-set period is denoted DHC. The definition of
lish relationships among the drilling variables. The approach DHC will vary with the operator, but will normally include the cost
follows four basic steps: of operations, plus overhead, base operations, incentive payments,
logging, transport, materials supply, marine vessel support, marine
Step 1. Select descriptor variables of the drilling process X1, ., supply base, port facility, and warehousing. The costs for comple-
Xp. tion and well test operations, production strings/liners, trees,
Step 2. Aggregate the descriptor variables as a component vector completion equipment, long-term daily rental of completions
according to category U, equipment, well design and programming, site survey and prepa-
ration, rig mobilization and demobilization, well suspension and
  re-entry cost, and plug and abandonment are usually not included
U¼ X1 ; .; Xp jAi  Xi  Bi ; in DHC. Total well cost TWC is dry hole costs plus all cost that were
where the values of Ai and Bi, i ¼ 1,., p, are user-defined. specifically excluded.

Step 3. Construct a regression model for given dependent variables 6.3. Source data
based on the sample data and aggregation category U,
The database employed in this study consists of 73 wells drilled
p in the Gulf of Mexico between 2002 and 2003. Well costs are not
X
4ðUÞ ¼ a00 þ a0i Xi ; normally made public by companies, and are considered confi-
i¼1 dential and proprietary. Data was compiled from two domestic
independents and one major who provided a representative
sample of their drilling program. Full data submission would be
preferable, of course, since it would avoid selection bias, but the
8
For example, underbalanced drilling accelerates drill time and may improve effort involved to collect and process ‘‘complete’’ data was outside
wellbore quality but requires specialized equipment and training for safe opera-
tions; 5th generation drilling rigs offer dual derrick capability to maximize drilling
efficiency but charge a higher rig rate; the use of specialized drilling fluids cost
9
more but can improve drilling efficiencies; the presence of shallow hazards can be For example, if management efforts are expected to be positively correlated
identified with special processing of seismic data but the extra cost of a directional with drilling exploratory wells, the coefficient associated with well type will be
well may not offset the delay of the hazard. underestimated.
1108 M.J. Kaiser / Energy 34 (2009) 1097–1112

the scope of the study. The collected data is meant to be illustrative, Functional relations evaluated at the component vector for well
rather than representative, and no claims regarding the generality w in category U are denoted as DHD(wjU), DHC(wjU), and
of the statistical relations to the GOM drilling environment should TWC(wjU). Performance metrics such as the rate of penetration ROP,
be inferred. If more comprehensive data was available, however, dry hole cost per day CPD, dry hole cost per meter DHCPM, and total
general models are readily established. All costs are reported in well cost per meter TWCPM, are derived from these measures. For
nominal dollars and are not inflation-adjusted. instance,
The technical characteristics of each well, such as drilled
interval, water depth, and casing sizes, and the data related to DI DHCðwjUÞ
ROPðwjUÞ ¼ ; CPDðwjUÞ ¼ ;
operational times is considered reliable, since these are well- DHDðwjUÞ DHDðwjUÞ
defined and frequently utilized industry measures. Data related to
interrupt times and cost is more subjective, due to the different DHCðwjUÞ TWCðwjUÞ
ways in which operators may define and calculate these values. We DHCPMðwjUÞ ¼ ; TWCPMðwjUÞ ¼ :
DI DI
consider measurement errors to be randomly distributed. An
attempt was made to ensure that all output metrics were properly
6.5. Expected signs
normalized, but again, constraints implicit in the data collection
place inherent limitations on the sample set.
The expected signs of the model coefficients are easy to ascer-
The restricted time window minimizes the impact of changes in
tain in most instances and provides a first check on the veracity of
reporting requirements, inflation, technological improvements,
the results. If the signs of the coefficients do not follow the expected
and volatility in the market rates of rigs. Only project data repre-
pattern and the variables are statistically significant, additional
senting actual cost was processed, and to minimize the influence of
structural issues may need to be investigated.
scale effects, batch and campaign drilling operations were excluded
Exploratory wells are expected to take more time to drill
from consideration. Only ‘‘new’’ wells are considered; sidetracks,
because of the potential risk involved, and since greater water
slot recovery/enhancement, and multilateral wells were excluded
depth means additional complexity, technical and environmental
from the sample.
complications, the water depth coefficient is also expected to be
All the wells in the sample set were drilled using dayrate
positive. The drilled interval is the length of the wellbore, and all
contracts with water/synthetic fluids, and none of the wells were
things being equal, it is clear that the greater the drilled interval the
horizontal or drilled in an underbalanced mode. Potentially useful
longer the time and the greater the cost of drilling. The drilled
characteristics such as hole size, volume removed, well status,
interval is expected to be a primary explanatory variable across all
maximum angle, and average mud weight were not available for
model types. The vertical interval is expected to be correlated to the
analysis.
drilled interval for exploratory wells, but for horizontal wells, the
correspondence is expected to be weaker.
6.4. Model specification The sign of the horizontal displacement, aspect ratio, and
extended reach ratio is uncertain. The aspect ratio and extended
The initial model reach ratio describe the geometry of the wellbore trajectory,
and since they are strongly correlated, only one of the variables
10
X should be included in the regression model. The coefficient of the
4ðUÞ ¼ a00 þ a0i Xi ; final bit size is expected to be positive since a large wellbore should
i¼1 take more time to drill and cost more than a small wellbore. The
is constructed for each of the dependent variables number of casing strings is a direct indicator of the complexity of
the well and the maximum mud weight is positively correlated
4ðUÞ ¼ fDHDðUÞ; DHCðUÞ; TWCðUÞg; with the pressure profile. Increasing the number of casing strings or
applying greater mud weight is expected to be associated with
for the input vector set,
slower drilling time and greater costs.
ðX1 ; .;X10 Þ ¼ ðWT; WD; DI; HD; VD; AR; ER; FBS; NS; MXMWÞ:
7. Drilling time and cost functionals
All the variables are numeric except well type which is cate-
gorical (WT ¼ 1, exploratory well; WT ¼ 0, development well). The 7.1. Statistical analysis
quantitative factors include water depth (WD, in 100 m), drilled
interval (DI, in 100 m), horizontal displacement (HD, in 100 m), Half of the drilled wells in the sample were exploratory and the
vertical depth (VD, in 100 m), aspect ratio (AR, unitless), extended other half development. Ideally, we would prefer to perform anal-
reach ratio (ER, unitless), final bit size (FBS, in inches), number of ysis on each class separately, but the small sample size precluded
casing strings (NS, integer valued), and maximum mud weight separate assessments. There was significant variation among the
(MXMW, in specific gravity). water depth and well trajectory characteristics, and most wells
Ideally, the input variables would be available for all aspects of required between 3 and 5 casing strings, which is typical in the
well construction in a consistent and comparable format, but for shallow waters of the GOM (Table 1). Most of the wells were drilled
practical purposes, we were selective in variable inclusion directionally which is common in development programs. The
because of data integrity and consistency issues. The coefficients drilled interval for the sample set ranged between 2000 and
ða00 ; a01 ; .; a010 Þ are estimated for each output measure. After the 4400 m. The average final bit size and maximum mud weight
initial model is computed and the statistical significance of the parameters are common in GOM shelf operations.
variables determined, the variable set is refined and the model Additional sample statistics presented in Table 1 include water
re-estimated. Usually, only variables with p-values < 0.05 are depth, drilled interval, horizontal displacement, vertical distance,
maintained, but other criteria may also be applied. The final re- aspect ratio and extended reach ratio. Most factors exhibited
estimated model is specified with all the regression coefficients significant variability and a low level of correlation. The extended
reported. reach and aspect ratio measures provide roughly the same
M.J. Kaiser / Energy 34 (2009) 1097–1112 1109

Table 1 Table 3
Wellbore statistics for Gulf of Mexico sample set (2002–2003). Drilling time and cost model results.

Metric (unit) Mean Std. deviation Min Max 4ðUÞ ¼ a0 þ a1 WT þ a2 WD þ a3 DI þ a4 HD þ a5 AR þ a6 NS


WT 0.51 0.50 0 1 DHD DHC TWC
WD (m) 1041 590 5 2432
a0 26.3 (3.8) 13.9 (4.7) 10.99 (3.0)
DI (m) 3254 1219 1118 7027
a1 1.79 (1.2)
HD (m) 823 789 0 2752
a2 0.19 (1.1) 0.46 (4.4) 0.60 (4.6)
VD (m) 3916 1365 1619 7758
a3 0.99 (7.6) 0.29 (4.7) 0.31 (4.1)
AR 0.36 0.39 0 1.5
a4 0.86 (1.8) 0.28 (1.3) 0.22 (*)
ER 1.1 0.2 1 1.7
a5 13.43 (1.4) 6.54 (1.5) 4.62 (1.8)
HL (m) 416 675 0 4540
a6 6.16 (3.7) 2.32 (3.1) 2.01 (2.1)
FBS (in) 10.6 2.1 6 14.8
R2 0.66 0.59 0.48
NS 4.1 0.9 2 6
n 73 73 73
MXMW (s.g.) 1.5 0.2 1.1 2.0
t-Statistics reported in parenthesis. (*) indicates t-statistic <1.
Based on 73 wells collected from 3 companies.

information about well geometry and should not both be incor- the variables and model fits appear reasonable considering the size
porated in the model. of the data set.
The average dry hole days, dry hole cost, and total well cost for
the sample set are depicted in Table 2. Derived measures such as 7.3. Example
the rate of penetration (ROP), dry hole cost per day (CPD), and dry
hole and total cost per meter (DHCPM, TWCPM) are also depicted. The average time and cost to drill an exploratory well (WT ¼ 1)
The average dry hole days per well was 31 and the average dry hole in the Gulf of Mexico in 220 m of water (WD ¼ 2.2) with a 1500-m
cost was $9.8 million per well. The average rate of penetration over drilling interval (DI ¼ 15), 250 m horizontal displacement
the drilling cycle was nearly 100 m/d. Observe that the standard (HD ¼ 2.5), 0.2 aspect ratio (AR ¼ 0.2) and four strings (NS ¼ 4) is
deviation of the performance statistics is in most cases only slightly estimated using the functional relations presented previously as
less than the average values, indicating that the data spread is follows: DHD ¼ 16.4 days, DHC ¼ $3.07 million, and TWC ¼ $3.68
significant. Total well cost is nearly $1000/m more than the dry hole million.
drilling cost. The dry hole and total well cost data are strongly
correlated, r(DHC, TWC) ¼ 0.93, with weaker but still significant 7.4. Time and cost relationships
correlations between dry hole days and cost: r(DHD, DHC) ¼ 0.75,
r(DHD, TWC) ¼ 0.72. The number of dry hole days to reach total depth is a primary
determinant of the dry hole cost, and similarly, the total well cost is
positively correlated with the dry hole cost. This relation should
7.2. Model results hold across all levels of aggregation, and if one quantity can be
predicted, then a functional relation could be used to derive the
Functional relations for DHD, DHC, and TWC are shown in other quantity values.
Table 3. The average dry hole days, dry hole cost, and total well The functional relation can be a simple multiplier, such as
cost are estimated as:
DHC
DHD ¼ 26:3 þ 0:2 WD þ 1:0 DI  0:9 HD þ 13:4 AR þ 6:2 NS;
kðUÞ ¼ ;
DHD

DHC ¼ 13:9 þ 1:8 WT þ 0:46 WD þ 0:3 DI  0:28 HD þ 6:5AR TWC


lðUÞ ¼ ;
þ 2:3NS; DHC
so that if the DHC(wjU) functional is established, then estimates for
TWC ¼ 11:0 þ 0:6 WD þ 0:3 DI  2:0 HD þ 4:6 AR þ 0:2 NS:
DHC(wjU) and TWC(wjU) follow from
Most of the variables are statistically significant and of the DHCðwjUÞ ¼ kðUÞDHDðwjUÞ;
expected sign, and the most significant variables for DHD include
drilled interval and number of strings; and for DHC and TWC, water
TWCðwjUÞ ¼ lðUÞDHCðwjUÞ:
depth, drilled interval, and number of strings. Drilled interval is
a primary variable in all three models. The statistical significance of Functional relations can also be derived directly, such as

DHCðUÞ ¼ a0 þ a1 DHDðUÞ;
Table 2
Performance statistics for Gulf of Mexico sample set (2002–2003). TWCðUÞ ¼ a0 þ a1 DHDðUÞ:
Metric (unit) Mean Std. deviation
Functional relations between dry hole cost and total well cost
DHD (days) 31.1 17.2 are depicted in Table 4. A positive and strong correlation exists
DHC ($ million) 9.8 6.1
TWC ($ million) 12.4 9.9
between cost and drilling days.
ROP (m/d) 99.3 66.8
CPD ($1000/d) 285.1 94.1 7.5. Example
DHCPM ($/m) 3097.8 1674.5
TWCPM ($/m) 4014.2 2545.3
Consider a well that is expected to take 55 days to reach total
DHC/DHD ($1000/d) 315.1 121.2 depth. From Table 4, the average dry hole and total well cost are
TWC/DHC 1.27 0.41 estimated to be DHC ¼ $16.8 million, TWC ¼ 2.02 þ 0.33(55) ¼ $20.2
Based on 73 wells collected from 3 companies. million.
1110 M.J. Kaiser / Energy 34 (2009) 1097–1112

Table 4 Appendix B. Drilling contract fundamentals


Dry hole cost and total well cost functional relations.

DHC (U) ¼ a0 þ a1DHD (U) TWC (U) ¼ a0 þ a1DHD (U) Drilling contractors
a0 0.29 (*) 2.02 (1.5)
a1 0.30 (9.4) 0.33 (8.6) The oil and gas industry is composed of operating companies,
R2 0.56 0.57 drilling contractors, and service and supply companies. In the U.S.,
n 73 73
operating companies are classified as independents and majors.
t-Statistics reported in parenthesis. (*) indicates t-statistic <1. Majors (e.g., ExxonMobil, Shell, Chevron) are ‘‘integrated’’ in the
sense that they produce, transport, refine, and sell petroleum
8. Conclusions products along the entire hydrocarbon supply chain. Independents
(e.g., Kerr McGee, Anadarko, Devon) are more specialized in the
Drilling operations are complex and many factors influence the sense that they only explore and produce (E&P) oil and gas. Inde-
time and cost to drill a well. In this paper, we described the factors pendents can range in size from just a few employees to several
that impact drilling performance, and outlined a general approach thousand, comparable to the majors. Outside the U.S., national oil
to understand how the factors of drilling can be quantified in companies play a prominent role in exploration and production and
a predictive model. The advantages and limitations of this approach are owned to varying degrees by the state.
were described as issues were encountered. We do not have perfect Majors and independents do not own the rigs that carry out
knowledge of the system and may miss important elements or drilling. In the past, they use to, but because of efficiencies,
interactions. The analytic tools introduce uncertainties, and since contractors who own and maintain a fleet of drilling rigs and
computational capabilities are limited, trade-offs and approxima- employ and train the personnel required to operate these rigs are
tions are necessary. Models also cannot anticipate the inherent hired by an operating company. There are many drilling contractors
variability associated with the human aspects of decision making. in the Gulf of Mexico, such as Pride International, Diamond
The model quantifies the nature of the drilling process and the Offshore, ENSCO International, Global SantaFe, TODCO, Transocean,
physical characteristics of the drilled well. It does not take into Noble, and Hercules Offshore. No drilling contractor is diverse
account estimates of the time or material costs for actions related to enough to perform all the work required to drill a well, and both
drilling. contractors and operators rely heavily on other companies to
It is not possible to identify all of the relevant characteristics of provide equipment, supplies and services. Service companies offer
drilling, especially when data sets are diverse and incomplete, but special support to the drilling operation such as mud logging, well
through the use of statistical analysis and empirical modeling, it is logging, completion, cementing, and catering.
possible to develop relatively robust relations that characterize
drilling performance through the specification of the primary Contract type
physical characteristics of the wellbore. Operators maintain
meticulous and detailed records of each well drilled, and by creating A footage contract is common onshore, where the contractor
an analytic framework that incorporates the main components of receives a specified amount of compensation for each foot of hole
drilling, it is possible to quantify the influence of individual factors drilled to a specified depth. A dayrate contract is common offshore,
and interaction terms. This generalized functional approach where the contractor charges on a dayrate basis until the well is
represents a stylized framework to understand the drilling process drilled. Combination contracts apply a footage rate to a certain
and is the best way to quantify performance evaluation. The basic depth and a dayrate below is sometimes used onshore. In a turnkey
structure of the analytic framework was illustrated using a small contract, a contractor agrees to drill a well for a fixed price to be
sample set which precluded detailed and generalized analysis. earned only when a specified depth is reached. Optional work after
turnkey depth such as logging, plug and abandonment operations,
and other services are usually performed on a dayrate basis. A
relatively new construct, incentive-based contracts, allow a rig’s
Appendix A. Drilling problem glossary dayrate to increase when it performs faster than the historic
average drilling curve. The basic idea is that an agreed upon norm is
General categories of drilling problems include: used to determine expected time and cost for the operation. Dril-
ling slower than the average time yields a financial penalty, while
 Fish or junk. Material that breaks (e.g., drill string, bit cone) or performance that beats the average results in a financial reward.
falls (e.g., pipe wrench) into the well. Contract durations can range from a few weeks, in the case of
 Stuck pipe. Drill string stuck in a well, frequently by doglegs or a single well, to as long as three to five years, for a series of deep-
an inappropriate mud program. water wells. Drilling contracts in shallow water tend to be written
 Sloughing shale. Shale that adsorbs water from the drilling on a per well basis, while in the deepwater, contracts are written
mud and expands when contracted by drilling mud, sloughing over a longer period of time. Turnkey contracts are used in the Gulf
(falling) into the hole. of Mexico in about 10% of wells drilled, and are significantly less
 Lost circulation. Excessive amount of mud lost during drilling, common in other basins.
usually occurring in a porous or highly fractured zone. The
sudden loss of fluid return causes the cuttings to be suspended Dayrate contracts
in the annulus and/or fall back down the well, clogging the drill
pipe. There are significant differences between dayrate and turnkey
 Formation (skin) damage. Mud filtrate which decreases the contracts with respect to risk allocation and insurance. Under
permeability of a reservoir rock near the wellbore. a dayrate contract, the E&P company pays the drilling contractor
 Embrittlement. Corrosive gas such as CO2 and H2S which a fixed fee per day worked. The responsibilities of the operator and
weakens the steel casing and drill string. contractor are specified in the drilling contract, but typically, the
 Abnormal high pressure. Hole pressure higher than the operator is responsible for: (1) rig relocation cost; (2) well design
expected hydrostatic pressure for that depth. and decisions regarding the drilling process, including total vertical
M.J. Kaiser / Energy 34 (2009) 1097–1112 1111

depth, mud type, mud weight, hole diameter, bit type, number of often demand a known drilling cost in order to justify project
casing strings, etc.; (3) coordinating supply purchases and delivery; approval, and co-ventures may want the operating partner to fix or
and (4) contracting for specialty services. guarantee the well cost. Although underbidding a project would
The contractor is responsible for (1) rig, including activities such allow a contractor to keep its rigs working in a downturn, if the
as tripping of drill pipe, running and retrieving riser, running contractor bid a price that did not cover its cost, significant losses
casing, and testing blowout preventer and surface equipment; (2) would be realized. Turnkey contracts offer strong incentives to
fully staffed crew, which include the toolpusher, driller, derrick- finish the job early since the contractor retains the full cost savings
man, several roughnecks and roustabouts, and other specialists of early completion. Turnkey bids can also be used as a way of
depending on the contract specification; (3) repair and mainte- benchmarking the performance of the drilling staff, since there
nance of the rig and equipment; and (4) specified materials and usually does not exist a way to properly validate a drilling AFE.
supplies. Circulating systems, horizontal displacement, weight on The main purpose of risk allocation in the offshore industry is
bit, and pump rate may be managed by the operator and contractor. to create a clear delineation of risk to enable each party to
The E&P company has one or more personnel stationed on-site measure the risk exposure it will absorb or insure. Risks are
to direct and supervise the drilling operation and provide support generally allocated without regard to cause, with the customary
services. Support personnel typically includes the company man, practice for the contractor to bear the risks of personal injury or
drilling engineer, mud engineer, geologist, and logger. Support and death of its personnel and assume liability for rig and associated
coordination tasks for the purchase and delivery of all other contractor equipment loss or damage. The operator normally
materials required for drilling, supplies, and equipment is borne by accepts liability for its own personnel and property, and in dayrate
the E&P company. Logging, core tests, drilling mud, drill pipe, bits, contracts, assumes responsibility for well related risks (including
casing and well equipment are typically supplied by the operator. pollution, wild well control, well damage or loss) and reservoir
The E&P company is also responsible for back-up services, trans- damage.
portation, and tugs during operations.
The contractor owns the rig and is responsible for its mainte-
nance, crew training, and management. The driller gives the orders References
in running the rig, but decisions on how the hole is to be drilled [1] 2007 Joint association survey on drilling costs. Washington, DC: American
are made by the E&P company. Drilling activities are monitored by Petroleum Institute; December 2006.
the company man, rig-based support personnel, and in some cases, [2] Bourgoyne AT, Millheim KK, Chenevert ME, Young FS. Applied drilling engi-
neering. Richardson, TX: Society of Petroleum Engineers; 2003.
the headquarters staff. Well data is analyzed, geophysical condi-
[3] Rose PR. Risk analysis and management of petroleum exploration ventures.
tions evaluated, and decisions are made as to the type of bit to AAPG Methods in Exploration Series No. 12. Tulsa, OK: The American
employ, its orientation, the weight and pressure of the mud, Association of Petroleum Geologists; 2001.
drilling tests, logs, cores to be performed, etc. [4] Baker R. A primer of oil well drilling. 4th ed. Austin, TX: Petroleum Extension
Service; 1979.
[5] Baum KE, Munson JR, Marshall JA, Vrooman DK. Turnkey drilling: bring cost
Turnkey contracts predictability to a risk business. IADC/SPE 39360. 1998 IADC/SPE Drilling
Conference, Dallas, TX; 3–6 March 1998.
[6] Marshall DW. The technical limit – illusion and reality. SPE/IADC 67819. SPE/
In a turnkey contract, the drilling contractor designs the well, IADC Drilling Conference, Amsterdam, The Netherlands; Feb 27–March 1,
purchases all the consumables, establishes contracts for the 2001.
transport and support services, drills the well, and charges the E&P [7] Bond DF, Scott PW, Page PE, Windham TM. Applying technical limit meth-
odology for step change in understanding and performance. SPE Drilling and
company a fixed sum of money for the operation. The role of the Completion Journal September 1998;13(5).
operator in this case is to specify the drilling targets and evaluation [8] Jones JA, Poupet P. Drilling the limit – a practical approach to breakthrough
requirements, and to specify the configuration of the completed performance. SPE 59207. IADC/SPE Drilling Conference, New Orleans, LA;
23–25 Feb 2000.
wellbore. Under a turnkey contract, the E&P company is selling the
[9] Sasifaldeen AJM, Thabet SAR. Integrate footage contract vs. day rate; drilling
project risk to a contractor who is paid a fixed fee in return for time and cost analysis. SPE/IADC 39255. SPE/IADC Middle East Drilling
drilling a well of given specification. The contractor assumes all Technology Conference, Bahrain; 23–25 November 1997.
[10] Kitchel BG, Moore SO, Banks WH, Borland BM. Probabilistic drilling cost
responsibility and is in total charge of drilling and completion
estimating. SPE 35990. Petroleum Computer Conference, Dallas, TX; 2–5 June
operations. The E&P company normally retains the risk for extreme 1996.
weather events and specific geologic formations, but if the [11] Hyne N. Nontechnical guide to petroleum geology, exploration, drilling, and
contractor expects the E&P company is letting the contract on production. Tulsa, OK: PennWell; 1995.
[12] Baker RA. A primer of offshore operations. 6th ed. Austin, TX: Petroleum
a ‘‘lemon’’ (i.e., extremely risky, difficult, and/or complex well), the Extension Service, The University of Texas; 2001.
contractor will price at a premium in a way that reflects higher [13] Pratt JA, Priest T, Castaneda D. Offshore pioneers. Houston, Texas: Gulf
expected cost. The ability to price a turnkey drilling contract to Publishing Company, Service; 1997.
[14] Leffer WL, Pattarozzi R, Sterling G. Deepwater petroleum exploration and
make a profit requires extensive experience, tight cost controls and production. Tulsa, OK: PennWell; 2003.
a balanced portfolio of projects to spread the risk. [15] Classification of fluid systems. World Oil June 2003;224(6):F3–6.
[16] Moomjian CA. Contractual insurance and risk allocation in the offshore drilling
industry. Drilling Contractor: January/February 1999, May/June 1999, March/
Negotiation and risk allocation April 1999.
[17] Pugh WW. The IADC offshore drilling contract. Oil and gas development on the
In a tight market, where rig utilization and dayrates are high, outer continental shelf, paper 8. Rocky Mountain Mineral Law Foundation;
1988.
contractors are able to negotiate favorable terms, while in a soft [18] Brett JF, Millheim KK. The drilling performance curve: a yardstick for judging
market, operators are often able to reshape contracts in their favor. drilling performance. SPE 15362, 61st Annual Technical Conference and
Thus, contractors tend to enjoy high rates, high utilization, and Exhibition of the Society of Petroleum, New Orleans, LA; 5–8 October 1986.
[19] 2000 Joint association survey on drilling costs. Washington, DC: American
favorable contract terms in good markets, and low rates, poor
Petroleum Institute; December 2001.
utilization, and unfavorable contracts in soft markets. [20] Jenkins PB, Crockford AL. Drilling costs. SPE 5266, SPE-European Meeting,
Under a dayrate contract, the dayrate charged and services London, England; April 1975.
performed are the negotiating factors, while under a turnkey, [21] Proehl TS. Pore pressures, fracture gradients, and drilling economics. IADC/SPE
27493, IADC/SPE Drilling Conference, Dallas, TX; 15–18 February 1994.
drillers compete on the price of the total job. The turnkey option is [22] Hariharan PR, Judge B. Benefits of dual-gradient methods for drilling upper
used primarily as a risk management tool. Marginal well economics hole sections. World Oil January 2003:44–7.
1112 M.J. Kaiser / Energy 34 (2009) 1097–1112

[23] Tibbits G, Tek T, Long RC, Miller BE, Black AD. World’s first benchmarking of [30] Holland D, Pritchard DM, Anand P. Seven drilling myths – managing successful
drilling mud-hammer performance at depth conditions. IADC/SPE Drilling drilling operations. Journal of Petroleum Technology 2003;55(9):44–8.
Conference, Dallas, TX; 26–28 February 2002. [31] Mannon DC. Reducing wells cost by increasing drilling efficiency. AADE 01-
[24] Hannegan DM. UBD tolls and technology solve conventional deepwater NC-HO-35, AADE National Drilling Technical Conference, Houston, TX; 27–29
drilling problems. AADE National Drilling Technical Conference, Houston, TX; March 2001.
27–29 March 2001. [32] Sanstran WC, Hawkins MJ. Perceiving drilling learning through visualization.
[25] Rowe SJ, Grittner S, Walters G, Shilling R, Brendling WJ. Using risk-based IADC/SPE 62759, IADC/SPE Asia Pacific Drilling Technology, Kuala Lumpur,
simulation models to assist in planning deepwater exploration and Malaysia; 11–13 September 2000.
production programs, west of Shetland and in the Gulf of Mexico. Proceed- [33] Schreuder JC, Sharpe PJ. Drilling the limit – a key to reduce well costs. SPE
ings of the 2000 Offshore Technology Conference, OTC 12952, Houston, TX; 57258, SPE Asia Pacific Improved Oil Recovery Conference, Kuala Lumpur,
May 2001. Malaysia; 25–26 October 1999.
[26] Dodson T, Dodson K, LeBlank M. Trouble time in offshore operations. Offshore [34] Ursem L-J, Williams JH, Pellerin NM, Kaminksi DH. The people aspects of
January 2004:47–9. drilling decision making in Real Time Operations Centers; SPE/IADC 79893
[27] Strom S, Balov K, Kjorholt H, Gasso R. The future drilling scenario. OTC SPE/IADC Drilling Conference, Amsterdam, The Netherlands; 19–21 February
19409, 2008 Offshore Technology Conference, Houston, TX, USA; 5–8 May 2003.
2008. [35] Liu X, Sinopec R, Samull R. ‘‘Actual 3D shape of wellbore trajectory: an
[28] Kaminiski DH, Pellerim NM, Williams JH. A new data integration and objective description for complex steered wells,’’ SPE 115714, SPE Annual
work process system for providing online real-time drilling collaboration. Technical Conference and Exhibition, Denver, CO 21–24; September 2008.
SPE 78343, SPE Europec Conference, Aberdeen, Scotland, UK; 23–25 [36] Amenta M. Identifying inefficient drilling conditions using drilling specific
October 2002. energy. SPE 116667, SPE Annual Technical Conference and Exhibition, Denver,
[29] Cayeux E, Genevois J-M, Crepin S, Thibeau S. Well planning quality CO; 21–24, September 2008.
improved using cooperation between drilling and geosciences. SPE 71331, [37] Gregory P. Improved drilling and completions decisions using advanced data
SPE Annual Technical Conference and Exhibition, New Orleans; September quality profiling techniques. SPE 115741, SPE Annual Technical Conference and
30–October 3, 2001. Exhibition, Denver, CO; 21–24, September 2008.