Electric-powered
hydraulic fracturing has been growing across the U.S. for a few years now.
Electric pumps can offer high horsepower and reliability. Now, in 2025, the
advent of fully automated hydraulic fracturing is in progress. An August 2025
article in Oil & Gas Journal by Alex Procyk gives some highlights of how AI
improves hydraulic fracturing.
Halliburton’s OCTIV Auto Frac
In January 2025, Halliburton announced their collaboration
with Coterra Energy on fully automated closed-loop hydraulic fracturing in the
Permian Basin. Halliburton utilized their OCTIV Auto Frac service as part of
their Zeus intelligent fracturing platform, along with its electronic pumps and
its Sensori fracture monitoring service. The initial rollout led to a 17%
increase in stage efficiency. There are several other advantages and
improvements that will be described below. The very significant improvements
will likely make auto fracs an industry standard, although it is not known how
fast this will happen. If the improvements are as good as they seem, then it
should happen pretty quickly. The OCTIV frac "ecosystem" is described below.
A January 2025 paper
presented at the SPE Hydraulic Fracturing Technology Conference and Exhibition
in Houston noted the advantages of automated fracs:
“Automating hydraulic fracturing enables operational
consistency in execution, reduces risk, and enables higher-level system control
for the operator. Automation also enhances employee utilization, freeing
resources for other tasks by eliminating tedious routines while pumping. In
2024, completion automation began to roll out across North America, and today,
controlling execution from the beginning to the end of stage, running across
numerous fleets.”
I was unable to access the
full paper, but an article in Drilling Contractor Magazine by Senior Editor
Stephen Whitfield explains the paper. According to Price Stark, Production
Enhancement Commercial Manager at Halliburton, and one of the paper’s authors:
“Any given second of a spread with 20 or so pumps could have hundreds of thousands of setpoints. And if you’re always fighting to find this optimal level of decision making, you’ll find that some people are very good at it, and some are not. We needed to put all of this decision making into a system that provides a base level of control and automation.”
The simple fact is that
manual control often results in less consistency in decision-making. This
results in less efficiency overall. Human operators differ by experience, risk
tolerance, response time, and judgment in critical situations. Automated
response does not have these limitations, but models must be trained properly.
The Drilling Contractor article notes that:
“…the operator determines the desired properties of the
fracturing fluid at the wellhead, such as slurry rate, treating temperature and
proppant concentration. The service company focuses on the operation of the
frac equipment, including setting the process parameters of each piece of
equipment.”
Automated decisions can be
manually overridden if necessary.
“Thousands of sensors installed on the equipment and
throughout the frac spread provide the Auto Frac system with real-time data on
the health of the pumps, the total load requested for a frac fleet, and how
best to distribute that load across the entire frac spread.”
Halliburton did extensive
testing of their Auto Frac through 2022 and 2023.
“Mr. Stark outlined four categories of metrics that
Halliburton sought to address while trialing the Auto Frac system: control,
consistency, performance and risk. Each was measured using conventional manual
frac control systems by the same crews in the same basins in the year prior to
deployment of the automated system.”
The wellhead slurry rate and
the rate of setpoint updates to each pump were analyzed to assess control. The
Auto Frac system allowed frac crews to modify pump rates more precisely.
Halliburton utilized the metric of hydraulic efficiency, a measure of actual
slurry volume divided by stage time, to evaluate frac efficiency. The results
showed a 4.6% increase in hydraulic efficiency, which is considered to be a
good number since increasing hydraulic efficiency beyond 1% has proved to be
difficult.
“Halliburton measured treatment performance during
testing as a function of time loss during the rate ramp period versus the
ideal, with “ideal” referring to the point where the wellhead rate equals the
maximum achievable wellhead rate within a given observed wellhead pressure. In
other words, how much slower does it take to reach the maximum achievable
wellhead rate than expected?”
The result was a 44%
improvement in auto frac compared to manual frac. Overall, Halliburton reports
that:
“…its automation technology executes up to 14,000 actions
during a single stage, reducing the workload on human operators by 88%, while
executing 2,720% more setpoints than human controllers.”
Evaluating risk was more
difficult due to the lack of established methods and metrics. As noted below,
they used a metric of “frequency of stages cut short” to evaluate risk, with
more stages cut short, usually screen outs, correlating to higher risk.
“For Auto Frac testing, Halliburton measured the
frequency of frac stages cut short. Mr Stark said it is common practice to
conclude a frac stage before the designed proppant mass is placed; this is a
response to perceived screenout risk. So, testing presumed that risk levels on
any stage cut short is higher than on stages pumped to completion, and a job
with lower frequency of stages cut short were presumed to be jobs executed with
less risk.”
The results show that with
manual fracs, there were, on average, 6.1 stages cut short, but with auto
fracs, there were just 1.6 stages cut short on average. Thus, improvements via
that metric are significant.
In June 2025, Halliburton
announced a collaboration with Chevron in Colorado that “enables
closed-loop, feedback-driven completions.” The companies developed
autonomous workflows that adjust well-completion behavior.
“Chevron’s work in closed-loop automation changes the
approach to hydraulic fracturing in shale and tight rock formations. Operations
can now react to a localized environment through real-time adaptation rather
than performance forecasting.”
Halliburton also notes that
their Octiv Auto Frac method optimizes equipment for extended life and
reliability. Their Sensori frac monitoring system allows them to measure the
fluid distribution in each frac stage in their near-well mode. They can also do
cross-well monitoring.
ProFrac Holding Corp. and Seismos Plan to Introduce
Closed-Loop Fracs Across All U.S. Basins
On August 18, 2025, ProFrac
Holding Corp. and Seismos announced their plan to introduce fully automated
closed-loop hydraulic fracturing across all U.S. oil & gas
basins. They plan to offer two modes, supervised and unsupervised.
Supervised mode allows human operators to make changes, while unsupervised mode
relies on automated decision-making. According to the announcement:
“ProFrac and Seismos are already working towards full
tech stack and crew-level integration. From day one, it has been designed to
scale across all fleets, enabling Closed Loop Fracturing for all supermajors
and leading independents.”
"Seismos is the innovator who introduced the
concept of Closed Loop Fracturing, and we are leading the market with the only
patented, transparent, fully-vetted technology that is years ahead of the
copycats," commented Panos Adamopoulos, CEO of Seismos. "True
closed-loop frac operations are built on the unbiased audit of frac
performance, a capability only Seismos provides. ProFrac's willingness to be
measured at this level speaks volumes about their commitment to performance and
transparency."
Seismos touts its measurement
while fracturing (MWF) capabilities. Their MWF QC system:
“…embeds a one-of-a-kind measurement referred to
as NFCI (Near Field Connectivity Index). NFCI is a patented measurement of the
reciprocal of flow resistance (NFCI = 1/flow resistance) in the near wellbore
area. The higher the NFCI, the higher the ability of the fracture network in
the near field (near wellbore) area to allow hydrocarbon inflow into the
wellbore.”
The NFCI metric is sensitive
to different parameters, including stimulation treatment, geology, stresses,
and the number of clusters taking fluid. A geology example is shown below that
optimizes treatment based on formation brittleness as derived from cores,
mechanical properties logs, and other petrophysical measurements.
Automated Pressure Management
E3 Company notes that there
is a clear shift in the industry toward automated frac equipment that is
basically unstoppable. The precision of automated frac tech is driving the
deployment. The introduction of smart controls, real-time data processing, and
responsive pressure management is leading to faster, safer, and more efficient
frac jobs. The use of embedded systems, programmable controls, and sensor
arrays to manage pressure, flow, and equipment behavior leads to better
performance. The use of AI/ML, IoT, and remote monitoring and control gives
additional improvements and the ability to make further improvements as more is
learned.
“It’s becoming increasingly clear that the future of frac
work is autonomous, data-driven, and remarkably efficient.”
In an article on their
website, E3 gives five common frac failures that can be prevented with
automated pressure control. I will summarize the five failures and how
automated pressure control prevents them.
1) Frac
Overpressure Failure Events – The automated system with
sensors takes hundreds of pressure measurements per second. This allows it to
open relief valves within milliseconds when the pressure gets dangerously high.
2) Slow or
Inaccurate Manual Bleed-Offs - The automated
system allows for depressurization, or bleed-off, to be triggered remotely with
precise control over rate and volume.
3) Stuck or
Failing Valves – The use of motorized or hydraulic remote valve
actuation that is built for high-cycle operation can result in faster detection
of valve problems. This can lead to making necessary adjustments before a
failure occurs.
4) Human
Error in High-Stress Conditions – E3 notes: “Actions
like opening a valve, relieving pressure, or calibrating a sensor are all
executed through a human-machine interface (HMI), reducing the chance of
misjudgment.” They note that their automation solutions are designed not to
support rather than replace frac operators.
5) Inconsistent
Response Times Across Shifts – people are
different and can often make different decisions based on similar data. In
other words, people tend to be inconsistent, especially compared to machines,
which can be programmed to be remarkably consistent.
Automated Hydraulic Fracturing Will Likely Become Industry
Standard
It seems very likely to me
that, for several reasons, automated fracs will become the industry standard as
the improvements are just too good to ignore. Unfortunately, this means fewer
people will be employed on frac crews in the future. That is the result of lots
of automation, and at some point, our societies will have to reckon with that.
However, many think that it won’t be a major issue. I do plan to research and
write about the effects of AI and automation on labor at some point in the
future.
References:
Frac
Bots, Coming to a Field Near You. Sarah Compton. Enspired by AAPG. August 26,
2025.
Coterra
Energy and Halliburton launch first fully automated hydraulic fracturing
program. Halliburton. January 6, 2025. Coterra
Energy and Halliburton launch first fully automated hydraulic fracturing
program
ProFrac
Holding Corp. and Seismos to Introduce Supervised and Unsupervised Closed Loop
Fracturing Across All U.S. Basins. August 18, 2025. TMCNET News. ProFrac Holding
Corp. and Seismos to Introduce Supervised and Unsupervised Closed Loop
Fracturing Across All U.S. Basins
Advancements
in Oil & Gas Operations: Exploring Automated Frac Equipment. E3 Company. Advancements
in Automated Frac Equipment
Top 5
Failures That Automated Pressure Control Can Prevent. E3 Company. August 19,
2025. Reducing
Failure With Automated Pressure Control
AI
improves fracturing performance. Alex Procyk. Oil & Gas Journal. August 14,
2025. AI
improves fracturing performance | Oil & Gas Journal
Chevron
and Halliburton enable intelligent hydraulic fracturing. Halliburton. June 12,
2025. Chevron
and Halliburton enable intelligent hydraulic fracturing
Halliburton
Debuts Automated Fracturing Tech in US Shale Fields Available to Purchase. Jaxon
Caines. Journal of Petroleum Technology. 77 (03): 28–31. Paper Number:
SPE-0325-0028-JPT. March 1, 2025. Halliburton
Debuts Automated Fracturing Tech in US Shale Fields | Journal of Petroleum
Technology | OnePetro
OCTIV®
digital fracturing services. Halliburton. OCTIV®
digital fracturing services
Auto
Frac: Improving Consistency and Control in Completions Available to Purchase. E.
Bogle and P. Stark. Paper presented at the SPE Hydraulic Fracturing Technology
Conference and Exhibition, The Woodlands, Texas, USA, February 2025. Paper
Number: SPE-223511-MS. Auto
Frac: Improving Consistency and Control in Completions | SPE Hydraulic
Fracturing Technology Conference and Exhibition | OnePetro
Sensori™
fracture monitoring service. Halliburton. Sensori™
fracture monitoring service
Testing
shows automated frac outperforms manual in control, consistency, performance,
risk: By replacing human decision making with data-driven controls, variations
in field crew experience that limit efficiency can be removed. Stephen
Whitfield, Senior Editor. Drilling Contractor Magazine. March 11, 2025. Testing
shows automated frac outperforms manual in control, consistency, performance,
risk - Drilling Contractor
A new
era of intelligent fracturing. Halliburton. Intelligent
fracturing
What's
MWF? Real-time Quality Control Agent for Fracturing. Seismos. Seismos Inc. | MWF™ - Real-time Fracturing
Quality Control Agent
Sensori™
Fracture Monitoring Service. Halliburton. Sensori_-_H014802
No comments:
Post a Comment