Artificial Lift That Thinks for Itself

Click here to listen to the Audio verison of this story!

In today’s data-driven oil field, having a system that manages and accurately interprets that data, in as close to real time as possible, is now a basic necessity. NOV and Lufkin are both using advanced systems to better manage production at various stages of a well’s life.

NOV’s Max Production Platform

Regardless of downhole pressure, Permian Basin production teams today are pressured from elsewhere—investors, stakeholders, etc.—to maximize investments by becoming ever more efficient in production, while keeping costs down.

Using digital systems for production management is becoming pretty much a necessity. Systems combining real-time monitoring, predictive analytics, and optimization tools are evolving quickly and are rapidly growing in use.

One such system is NOV’s Max Production platform, part of their larger Max™ IIoT platform. It’s designed specifically to maximize artificial lift production operations.

 

Balancing the Trade-Offs

Operator success requires standing on a three-legged stool: maximizing output, extending run life, and operating with maximum efficiency—requirements that often compete. From the outside they deal with grid/electrical cost issues and other expenses.

“Whether you’re focused on volumes, run life, or power draw, the pain points are converging,” said Senior Director of Automation, Control, and Optimization at NOV Austin Wilcox. “We see operators pushed to make daily decisions without the right information. It’s not necessarily a lack of data—they’re just not getting the right insight at the right moment to make the most informed decisions.”

That lack of insights can lead to suboptimal pump sizing, insufficient failure detection, or insufficient setpoint adjustments. Smarter and more adaptive systems can reduce costly interventions and improve production.

 

Adaptable Tools Leverage Immediate Data

NOV says Max Data combines edge-to-cloud architecture and a package of modular application layers to provide actionable data. It provides real-time monitoring, analytics, and control in one package for all forms of artificial list and many other production assets.

It’s all about usability, said Stephen Peters, director of user experience for NOV’s Digital Services. “We’ve prioritized the experience for the people who actually use the system day to day. Whether it’s an operator in the field or an engineer in a central office, Max Production delivers consistent, customizable views that align with their role and workflow. That means faster response times and better collaboration.”

Scalable edge hardware—located onsite where the datapoints are being collected—along with dashboards, collect and contextualize the data immediately, regardless of the cloud connectivity status at any given moment.

 

Integrated Lift Planning and Production Management

Looking for a way to expand their approach to production management, NOV in 2024 acquired Extract, a provider of ESP and H-pump technology. A key part of that acquisition included Extract’s proprietary Xsize platform, created for production forecasting and optimization across all lift systems. NOV integrated that into Max Production, where it applies nodal analysis, comparative lift modeling, and pump life cycle assessments to help optimize production across a well’s productive life.

“Xsize was originally developed by engineers, for engineers,” said Michael Owen, product line director of Data and Optimization Services at NOV. “Its integration with Max Production allows teams to simulate and optimize their entire lift strategy—from sizing to real-time adjustments—using a common data foundation. That means we’re not just reacting to problems, we’re preventing them.”

 

The Autonomous Future

Harnessing, processing, and contextualizing the increasing complexity of operational data and giving it a voice in decision making is a driving factor in the rush toward AI analysis. The more decisions a system can make independently of human intervention, the faster and more efficient that system is. Said Peters: “Edge technology will be crucial to this transformation. Our edge platform is built to run advanced analytics and AI models locally, even when the network is down. That kind of resilience is essential.”

We’re not there yet, but momentum is building, Wilcox pointed out. “NOV is actively working to integrate the capabilities required to support this evolution through its Max Production platform.” The goal involves progressing from reactive to predictive strategies.

Owen added, “Getting autonomous is not just about technology—it’s about trust. Operators will need time and experience with these systems in the field to build the confidence required for broader deployment.”

Timely context is the key, said Wilcox. “Ultimately, it’s about giving teams the context they need to act—whether it’s selecting the right lift method or tweaking performance in real time. As operations get more complex, the simplicity of the user experience and the power of the underlying analytics are what really drive value.”

 

Lufkin Revolutionizes 40-year-old Algorithm

Algorithms that are used to optimize rod lift production worked very well from their inception in the 1960s and all through the vertical well era. But the Shale Revolution’s long laterals and deviated wellbores have created new issues with resistance from rod length and friction against the sides of the well that have rendered those algorithms obsolete. They don’t give producers accurate information about pump operation.

That’s a problem because, without proper data, producers might experience premature failures due to over-pumping or might defer production due to under-pumping.

LUFKIN Well Manager™ (LWM) 2.0 with NOVAWAVE™ Advanced Wave Equation Technology, released in April, can improve downhole diagnostic accuracy, leading to more precise pump fillage estimation and better operational control decisions, said Peter Westerkamp, VP of automation for Lufkin Industries.

 

The Issue with How We’ve Always Done It

Many shale wells start production either by flowing or with an ESP, due to the latter’s ability to produce the large volumes common early in a shale well’s productive life. As production drops to 300-700 barrels per day (bpd) or less, other options become more efficient, and the iconic rod pump is one of them. As the Permian Basin’s revival has been based on quickly depleting shale wells, Westerkamp noted that Basin production could benefit from an updated wave equation.

Wave-equation algorithms designed to assess a carefully drilled vertical well were developed in the 1960s, when vertical wells were the only option. Modern operators drill multiple wells from a single pad to exploit different target zones. This can result in wellbores that include rapid directional changes and long, inclined sections. That situation affects completions and production no matter the lift system, but it’s particularly challenging for the long rod strings involved in rod pumping.

Part of the issue is with data—rod pumps historically have not had the downhole sensors available to ESPs and gas lift, so there’s no immediate feedback data. And today many rod pumps are being installed below the kickoff point into the horizontal. That, along with the deviated vertical wellbores, will cause the rod string to rub against the tubing, creating friction and inaccurate pump cards.

 

Tooling Up

So, said Westerkamp, about 10 years ago Lufkin engineers realized accurate rod pump data would require downhole sensors, such as were common to ESPs and other lift systems. So, they began to develop those for rod pumps. The end result? “We refer to those as downhole dynamometer tools or downhole load cells and they become part of the rod string,” Westerkamp said.

Installed in multiple places along the string, these memory tool sensors measure load acceleration, pressure, and vibration. After allowing them to collect data for several weeks, Lufkin analysts retrieve the tools and process their data.

“We then used that information to develop a completely new version of the wave equation,” Westerkamp said. He added that the company partnered with Texas Tech’s mathematics department to create the updated equation.

The company began testing with these tools in 2016. The 2025 release of LWM 2.0 included NOVAWAVE, the wave equation that resulted from those years of collecting and analyzing the downhole data.

 

Advantages: Accurate Pump Fillage

To maximize profitability by producing efficiently and minimizing damage, rod-pumped shale wells often use a rod pump controller (RPC) or variable speed drive (VSD) rather than run at one speed all the time. But to accurately control the VSD’s pump speed or the RPC’s start-stop intervals the controller must accurately measure pump fillage.

If the producer is underestimating the fill rate, it’s slowing production and reducing the crude available for sale. But if the fill rate is slower than anticipated, it could create fluid pound, damaging the rod string, the pump, and other components—increasing costly workovers and replacements. And during a workover, production stops.

Westerkamp says NOVAWAVE can prevent those issues by using accurate downhole data to operate the pump. “With this wave equation update we can more accurately determine what the pump fillage is, so we can avoid over pumping a well.”

More accurate cards will help determine gas interference, pump spacing and improve the PIP calculation, all of which will assist in better well operations

NOVAWAVE’s system may also just maintain the same production with fewer strokes. “If we can reduce pumping time by, for example, 5 percent and not lose production, we expect that that rod string is going to last 5 percent longer,” Westerkamp said.

 

Paul Wiseman is a longtime writer in the energy industry.