SPE/IADC-173148

Well Cost Estimation and Control

Advanced Methodologies for Effective Well Cost
Management

Presenter: John de Wardt

DE WARDT AND COMPANY
 Slide 2

Significant improvement in well cost estimation and control

is required

• Effective well cost estimation and control is critical

• Many poor practices across drilling industry
• Classification of costs provides a foundation
• Linkage to schedule, uncertainty and risks drives the outcome
• Probabilistic estimation has value typically not seen
• Effective well cost estimation and control workflow
• Time to do it right
 Drilling & Completion Cost Estimation and Control Framework

SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt

 Drilling & Completion Cost Estimation and Control Framework Slide 3
Class 5 – Concept Screening Costs sufficient to enable early
Level 1 High Level Large range decision making
Project Milestones from Range is large.
Analogous Data

Class 4 – Feasibility Level 2 Wide accuracy,

Preliminary Budget. Major schedule Broad Range development of cost
Project Summary
& range character for
Schedule
Cost Impact Simple cost ranges decision making

Probability
Scheduling Risk Analysis
Class 3– Budget Authorization Tightest accuracy available from the
1st Control Estimate Level 3 Detailed line item Increased
design of the wells and definition of the
Integrated Project Analysis and Granularity
execution.
Schedule mitigation Characterization from probability
S curves analysis defines the accuracy for
Tornado diagrams decision making
P10 / P50 / P90
Distributions
Class 2 – AFE Defines the cost against
Detailed Control Baseline Level 4 Improved which the drilling operation
Mitigation
Functional Confidence in will be measured
& B
Intermediate values
Management Plans
Schedule Developed o
Final
t Distributions
t
Class 1 – Detailed Control o Highly detailed cost tracking to provide
manager with continuous detailed status
Level 5 m against plan
Activate
Working Level Contingencies Reference for invoice reconciliation
Schedules
u
p
SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt
 Slide 4

Why does the industry need to improve?

• Well feasibility costs drive the economic analysis

− leading to a decision to implement a project.
• Drilling budget and AFE define a significant
component of the expenditure of an operating
company.
• Actual well costs define the expenditure made
− provide reconciliation to the invoiced
quantities.

SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt

 Slide 5

Poor practices are prevalent

• Deterministic estimates used when uncertainty and risk prevail

• Wrong people make the estimate
• Estimators lack a defined system and training
• Contingency is a fixed percentage (often 10%)
• Probabilistic estimating is misapplied through lack of understanding
• AFE variances after overruns occurred

SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt

 Slide 6

Cost classes follow a standard

Estimate Project End Usage Methodology Expected
• Defined by AACEI and DOE Class
(definition)
Definition Accuracy range

Very low Concept Screening Analogy, judgment, factoring, +30 to +100%

• Logical stages that match well Class 5
Limited effort
Quick preparation
modeling -20 to - 50%

design / programming (0 – 2%)

• Creates consistency with facilities Class 4
Low Feasibility / study
Preliminary budget approval /
Probabilistic +20 to +50%
-15 to -30%
gate approval
and FDP
(2 – 15%)
Medium Budget authorization or Semi detailed unit cost Probabilistic for +10 to +30%
control risk / uncertainties -10 to -20%
Class 3 Support full funding.
1st control estimate

(10 – 40%)
High AFE Detailed unit cost +5 to +20%
Detailed control baseline Probabilistic for risk / uncertainties -5 to -15%
Class 2 Expenditures will be reported
against this

(30 – 70%)
Complete Detailed Control Detailed unit cost +3 to +15%
Current control estimate Deterministic with contingencies -3 to -10%
Class 1

(50 – 100%)

SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt

 Slide 7

Probabilistic estimation is a powerful tool

• Most effective method to incorporate uncertainties & risks

− quantified
− graphically displayed
• Transparent method to view through class development
• P90 provides a logical contingency funding level
• Methodologies exist to correctly apply Monte Carlo simulation
− Distribution types for time / cost variances
− Roll up to limited summed activities / costs

SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt

 Slide 8

“S” Curve development for a well

Not So Risky vs. Risky

• View the effects of 1.0

uncertainties 0.8
Base well cost estimate
• View the effects of considering uncertainties
risks 0.6 in costs and activities.

Cumulative Probability
Base Contingency due to risks and
• Define the Contingency impacts

contingency based 0.4

on transparent 0.2
Total Contingency summed from identified contributors.

logic

Mean = $10,849.00
0.0

Well Cost (MM)

000

500

000

500

000

500
Base well cost estimate
adding risk events and
impacts.

SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt

 Slide 9

Maturing the cost estimate effect on “S” curve

1.0

• Maturing the cost P90 Maturing

estimate alters the 0.8 Probabilistic estimate
matures, curves become
probability profile steeper:
• Later cost estimates 0.6
- less risk

Cumulative Probability
- Less uncertainty. Most likely well cost ranges reduced
ought to fit within as risks and uncertainties are
0.4

earlier cost estimates managed during the planning process:

- Mean
unless the scope has 0.2 - P10/P90
changed P10 Maturing
0.0

Well Cost Estimate (MM$)

50
20

30

70
60

80

90

100
40
SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt
 Slide 10

Tornado diagrams rank the sources of uncertainty

Estimated Well Cost (MM$)

• Visually identify most Correlation Coefficients (Spearman Rank)

0.73

influence 0.40

Most Influence
• Identify opportunities 0.21
0.33

to invest to reduce 0.19

Tornado diagram
0.16

0.15

0.13
prioritizes
A
0.10 uncertainties and risks
0.09 based on their
0.08 hierarchical influence
0.07
on the well cost.
0.06

0.05

d
0.05

0.04

Coefficient Value

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8
-0.1

SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt

 Slide 11

Well cost estimation and control can be improved significantly

through adopting an effective workflow

• Design a suitable well cost template

− Activity based costing
• Define responsibility – Drilling Department
• Adopt the cost classes
• Link to schedule and risk management
• Distinguish Exploration / Appraisal from Development well cost cycle

SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt

 Drilling & Completion Cost Estimation and Control Framework Slide 12
Class 5 – Concept Screening Costs sufficient to enable early
Level 1 High Level Large range decision making
Project Milestones from Range is large.
Analogous Data

Class 4 – Feasibility Wide accuracy,

Preliminary Budget. Level 2 development of cost
Major schedule Broad Range
Project Summary range character for
&
Schedule decision making
Cost Impact Simple cost ranges

Probability
Scheduling Risk Analysis
Class 3– Budget Authorization Tightest accuracy available from the
1st Control Estimate Level 3 Detailed line item Increased
design of the wells and definition of the
Integrated Project Analysis and Granularity
execution.
Schedule mitigation Characterization from probability
S curves analysis defines the accuracy for
Tornado diagrams decision making
P10 / P50 / P90
Distributions
Class 2 – AFE Defines the cost against
Detailed Control Baseline Level 4 Improved which the drilling operation
Mitigation
Functional Confidence in will be measured
& B
Intermediate values
Management Plans
Schedule Developed o
Final
t Distributions
t
Class 1 – Detailed Control o Highly detailed cost tracking to provide
manager with continuous detailed status
Level 5 m against plan
Activate
Working Level Contingencies Reference for invoice reconciliation
Schedules
u
p
SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt
 Slide 13

Recommendations and Conclusions

• Develop a well cost estimation and control process based on best

practice
• Follow the class levels
• Integrate schedule and risk into cost estimation
• Assign competent resources
• Apply probabilistic methods in a competent manner
• The cost ranges are the ranges – do not make them up

SPE/IADC 173148 – Well Cost Estimation and Control – John de Wardt

 Slide 14

Acknowledgements
Co-Author: Susan Peterson, Risked Decisions

Thank You / Questions

John de Wardt