This document discusses lessons learned in how not to implement drilling automation. It begins by defining drilling automation and providing examples of automated processes like maintaining downhole weight on bit and optimizing mechanical specific energy. It then outlines common mistakes made in automation projects like not including drillers, having insufficient data quality, and giving drillers too many or too few controls. The key lessons are that automation must improve performance, drillers must be central to the design and implementation, reliable data and controls are essential, and human factors like training and complacency must be addressed. Critical success factors for automation include deciding what to automate and implementing with consideration of both technical and people issues.

1. Society of Petroleum Engineers
Distinguished Lecturer Program
www.spe.org/dl 1
Dr. William L. Koederitz, SPE, PE
Lessons Learned,
How NOT to Do Drilling Automation

2. Outline
• What is drilling
automation?
– Examples
– Pros and Cons
• How NOT to do drilling
automation
– A positive side will also be
shown!
• Conclusions
2

3. Drilling Automation
• The technique of operating or controlling
a process by highly automatic means,
reducing human intervention to a
minimum.
• Mechanization refers to the replacement
of human power with mechanical power of
some form.
3

4. The 10 Stages of Automation
4
Level Automation Description
10
The computer decides everything, acts autonomously, ignoring the
human.
9 Informs the human only if it, the computer, decides to
8 Informs the human only if asked, or
7 Executes automatically, then necessarily informs the human, and
6
Allows the human a restricted time to veto before automatic
execution, or
5 Executes that suggestion if the human approves, or
4 Suggests one alternative
3 Narrows the selection down to a few, or
2 The computer offers a complete set of decision/ action alternatives, or
1
The computer offers no assistance: human must take all decisions
and actions
IEEE Transactions on Systems, Man, and Cybernetics- Part A: Systems and Humans, Vol. 30, No. 3, May 2000

5. Example – DWOB Control
• DWOB = “Downhole Weight on Bit”
• SWOB = “Surface Weight on Bit”
• DWOB ≠ SWOB
• Constant DWOB provides better results
– Higher Rate of Penetration
– Better directional control
5
SurfaceWeight
W eig ht o n B it
NormalForce

6. Manual DWOB Control
• Control process by driller
– Read slow-speed DWOB
– Compare to desired DWOB
– Adjust SWOB setpoint in autodriller
• Holds DWOB “close” to desired
• Requires constant monitoring, adjusting
• If downhole conditions change, must react
rapidly
6

7. Automated DWOB Control
• Driller sets bounds on DWOB, SWOB
• Automated optimization process
– Analyze high-speed surface and downhole
drilling data
– Compute change in SWOB
– New SWOB sent direct to rig
• Driller now only has to monitor
• Holds DWOB very close to desired
• Reacts quickly to changes downhole
7

8. Example – MSE Optimization
• MSE = “Mechanical Specific Energy”
• MSE = energy in / volume of rock drilled
• Lower MSE  more efficient drilling
8

9. Manual MSE Optimization
• Optimization process by driller
– Change Bit Weight and/or RPM
– MSE response dictates next change
• Performance improvement
– More as driller gains experience
• Requires constant monitoring, adjusting
9

10. Automated MSE Optimization
• Driller sets bounds on Bit Weight, RPM
• Automated optimization process
– Analyze recent drilling & MSE data
– Search technique selects Bit Weight, RPM
– New Bit Weight, RPM sent direct to rig
• Driller now only has to monitor
• Performance improved in most cases
– Can’t compete with dedicated expert driller
10

11. Why Automate?
• Efficiency
– Tasks that are repetitive and require
continuous monitoring can be done more
consistently with automation.
– Free up rig crew for other tasks
• Enhance Crew Capability
– Shortage of experienced individuals at the rig
• Improved Performance
– Do things that people can’t do (non-stop)
• Safety
11

12. Risks of Automation
• Complacency
• Loss of ownership
• Dependent on data & control
quality
• Maximum performance limited by
“smartness” of automation logic
– In the specific situation
• Automation can not innovate
– Only motivated people can do that
12

13. When & What to Automate
• Selection Methods
– Look for good automation applications
– Look for performance improvement
opportunities
• Define automated and non-automated options
• Decide based on your criteria
– Return on Investment
– Safety
13

14. Drilling Automation in SPE
• SPE DSATS
– Drilling Systems Automation Technical Section
– Purpose is to accelerate automation in drilling
– On SPE website, workshops, forums, …
– SPE/IADC-173010-MS “Drilling Systems Roadmap
– The Means to Accelerate Adoption”
• IADC ART
– Advanced Rig Technology Committee
– Focused on safety and efficiency of automation
14

15. How Not to …
“The office saw value and
wanted it, so the rig will too.”
•Performance-motivated rig
•Office often out of touch with actual
rig operations
– Rig crew sees the negatives and
focuses on them
•Solution
– Include driller from the start
– Change how people work
15
Aha!!!

16. How Not to …
“The office saw value and wanted it, so the rig
will too.”
•NOT a performance-motivated rig
•Solution
– Change to performance-motivated rig!
– If not willing to do that:
◦ Acceptance will be an issue
◦ Design in value that has meaning at rigsite
–Make their life easier
16

17. How Not to …
“Driller is no longer needed. ”
•Driller is the core of rig activity
•If he feels left out, automation will not work
– Even if no action is required on his part
•Solution
‒ Design system with driller at center and in
control
‒ Treat driller as most-critical automation enabler
17

18. How Not to …
“That rig’s data was good enough for drilling,
so it’ll be fine for automation.”
•Typical rig data is never good enough
– Often already insufficient (if you really look)
•Reliable, high-quality data is a must-have
•Solution
– Investigate rig data quality, upgrade as needed
– Continuous monitoring of data quality
18

19. How Not to …
“That rig’s controls were good enough for
drilling, so they’ll be fine for automation.”
•Reliable, sufficiently precise control of rig equipment
is a must-have
•Typical rig control is often not precise enough or is
not readily accessible
•Solution
– Evaluate rig control capability, resolve issues
– Continuous monitoring of control quality
19

20. How Not to …
“Since it’s automated, driller only needs to turn
it on, not understand how it works.”
•This reduces effective use (loss of value)
– Worst case, destroys rigsite acceptance
•Optimum use by rig  maximum value
•Solution
– Design so driller is well informed of how it works
– Enhance comfort level (simulator exercises a +)
20

21. How Not to …
“This rig is a sister rig to the last one we
automated, so we are ready to go.”
21
• Every rig has some unique aspects
• Office records often aren’t perfect
• Solution
‒ Do a detailed rig survey
‒ Build configuration specific to rig
‒ Pre-test configuration in lab

22. How Not to …
“It’s a highly-automated system, so there
shouldn’t be any maintenance for the rig to
do.”
•Maintenance needed for optimum, safe
performance
•Changes in rig, sensors, drilling, …
•Solution
– Design for easy, minimal maintenance
◦ Automated diagnostics or remote monitoring
22

23. How Not to …
“Their only choice is on or off.”
“Let’s let them adjust everything.”
•There is an optimum level of interaction for
each driller and situation
•But too many levels are confusing
•Solution
– Analyze drillers, identify group(s)
– Design for some variation in drillers
◦ Basic vs advanced
23

24. How Not to …
“Automation seems to be going well, so
driller must be paying close attention.”
•Complacency is a risk
– The “better” the automation does its job, the
higher the risk
– A tough problem to solve
•Solution
– Human factors engineering, in some form
24

25. How Not to …
“Let’s make the system do everything (we
think) they need. They’ll sort it out.”
•The driller is over-loaded by this, resulting in
misuse or non-use
•Solution
– Design the system as a suite of tools
◦ Driller picks the right tool for the right job
– Key decision criteria are simplicity,
modularity, benefit/cost ratio
25

26. Conclusions
• Automation is a tool to improve
performance
– Pros and cons, per application
• Critical success factors
– Deciding if and what to automate
– Design and implementation
◦ People issues often > technical issues
◦ Do not leave the driller out!
26

27. Society of Petroleum Engineers
Distinguished Lecturer Program
www.spe.org/dl 27
Your Feedback is Important
Enter your section in the DL Evaluation Contest by
completing the evaluation form for this presentation
Visit SPE.org/dl