From the drill bit to the back office, information technology’s efficiencies and insights are helping the oil and gas industry survive and even thrive in more and more new ways every day.
Case in point: a line on the Erdos Miller website points out that, 10 years ago, the words “technology” and “completions” would not have been used in the same sentence—but now they are quite compatible. It’s a sign of how far information technology is reaching throughout the industry, improving efficiency and accuracy wherever it goes.
Their website also describes the company, founded in 2009, as “an engineering and product development firm with a focus on upstream oil and gas.” Their products include the Eclipse Touch MWD (measure-while-drilling) Surface system, along with directional drilling controllers and a free drilling analysis program called WAVE.
Company Cofounder and President Kenneth Miller said that the importance of Big Data is becoming obvious to more and more oil patch companies. While this understanding has been mostly growing on the production side, it’s becoming a key to equipment design as well.
Who needs data? According to Miller, “Anyone designing any sort of product.” He adds, “It used to be, put the engineers together, design the product, ship it, [make a] profit.”
In the current world, Miller described a process in which the engineers still initiate the design, but the early product must collect performance data that can inform future improvements.
“So this continuous data cycle is critical,” Miller said. Whether a company is designing drilling rigs or anything else, “If you’re not collecting every bit of analytic about your product’s performance that you can, then acting on it, you are not going to be relevant in the future. That’s just going to be how the world goes about things.”
Miller is passionate about how IT is going to shape the future—and the market of today.
“I can’t tell you how many times I chuckle to myself when they say, ‘Oh, mud motors are going to continue to be better than rotary steerables.’ I’m like, no, come on, that’s like the next wave of technology. You’re just kidding yourself if you think we’re going to make the technology we were drilling on 50 years ago better than the technology that’s coming out today.” It may take time for new technology to take market share away from the older versions, but “I’m sorry, the horse and buggy are going to go away and the car’s going to take over.”
Simply collecting relevant data is not enough—that data must first be tamed, then used for instruction. The flood of data can be overwhelming to the unprepared. “People severely underestimate the amount of effort that goes into actually extracting value from that data.”
He sees the need for a team of 3-5 data engineers whose job is to make sense of the data—“to write scripts and find different ways to slice the data and statistically analyze it and feed it into analytics platforms” to learn how the tool is performing and how to make improvements. These are things that only humans can do—no Machine Learning (ML) or Artificial Intelligence (AI) can “think” of how to look at the data.
“We’re talking about a million bucks a year just to make sense of your data,” he said. This depends on the size of the platform and how it’s being done.
Erdos Millers’s own specialty, MWD systems, have changed drastically in the last 10 years. Early tools came with 24 MB of memory, which was a small amount even then, limited by the extreme downhole conditions in which the tool had to survive.
Their new MWD surface software, known as Eclipse Touch, provides mud pulse decoding. It includes a gigabyte of memory for the BHA tools, which informs operators about ways it operates in action, along with data about the well. That’s 3-4 gigabytes per week for each job, and with 30 jobs going the number of datapoints becomes astronomical.
With all that information flooding in, Miller says, “I still feel this data is so underutilized—there’s still so much more we could study to it. What’s really scary to me is that we’ve already become so efficient as an industry at producing oil, that if we really did utilize all this data, the amount of oil we could output would be a little bit terrifying.”
Already, the company’s data engineers have programmed the system to run certain experiments on its own to find the best procedures. Those that boost results, it keeps, and those that don’t improve things are discarded. This way, it’s making improvements while engineers sleep.
“It’s a fantastically cool time to be in engineering.”
In the back office, some “datapoints,” if you will, have been manually created for more than 100 years. Field tickets written by hand and delivered by hand—when they weren’t lost or obliterated by coffee stains—were the way of life until the advent of computers. Now several companies, including Greasebook, are turning those manual entries into electronic ones, boosting efficiency and accuracy in the process.
Founder and CEO Greg Archbald grew up in a family–owned oil and gas accounting firm whose clients were both operators and service companies in and around Oklahoma City. As he worked in the family business, he saw clients having difficulty with field-to-office communications. He was baffled at how many companies tolerated such an important and costly situation.
“There were a lot of misunderstandings, reports came in late, lost tickets, paper gauge sheets. That stuff would get phoned in, texted in, faxed in, emailed in by five, eight, ten different pumpers, and they’re all sending it to you in a different format. You’re going mad trying to collect all this stuff just to get it into something workable, so you can actually start making heads or tails of the thing,” he recalled.
The idea that would become Greasebook came about in 2009 or 2010, after the advent of the iPhone. “So we said, ‘What a great idea—wouldn’t this work well in the oilfield?’”
Due to the the oil patch’s famed resistance to technology adoption, Greasebook originally met with doubts about whether pumpers would use it. Archbald was told, “Most pumpers are mechanics, they’re mechanics first, they’re very good at what they do—they’re not the most tech-savvy crowd.”
The biggest concern, he heard, was that the best pumpers were long-experienced people who could take one look or one listen at a pump site and instantly know whether everything was working right or not. Operators were worried about running those people off with new technology.
“Fortunately, Apple [iPhone] and Google [Android] fought that battle for us, for better or for worse,” he said with a laugh. “Now everybody has a smart phone device, even guys that are 80 years old in the field and still working, because they want to keep up with their grandkids.”
Possibly a bigger part of the resistance, Archbald believes, came from pumpers wary of new systems because they’d been burned before by software that was too complex to help. Previous software actually added 1-2 hours of work after a long and challenging day.
With Greasebook, on the other hand, “It is very simple, and pumpers generally find they save 30 minutes a day in paperwork. It can be a win-win for the pumper and the operator as well,” Archbald said.
Data entered into a smart device is instantly available for reports to accounting and supervisors, who can tell where the pumper was and at what time.
One thing pushing greater acceptance of efficiency software like Greasebook, said Archbald, is the COVID-19 work-from-home edicts. “I think it’s really accelerated things people were talking about for a long time,” he said. “The digital oil field—which I’d been reading about for 25 years—it’s always been this mirage on the horizon.” It has been seemingly out of reach or not quite the right time. All that changed in 2020.
Many of the changes are permanent. “Never going back,” he declared, “It’s a new day. The Great Crew Change was another theme we’ve been talking about for 15 years,” he added, citing the trend of older workers entering retirement and being replaced by a new generation more open to technology. Operators over the last six months have told him that’s happening now as well.
With many older systems, Archbald said, pumpers would bring their paperwork home to their wives to handle. So Archbald has received voicemails from wives saying, “Now I get to talk to my husband,” or “We’re not stuck Saturday afternoons doing paperwork—we can go do activities together.”
Another back office software company is W Energy Software (formerly Waterfield Energy). They were founded in 2009 in Tulsa.
Pete Waldroop, the company’s CEO, also noted that older oil and gas software was less user-friendly than today’s versions. Waterfield began creating his company’s system with the idea of replacing Microsoft Excel, but quickly changed course. “What we needed to do was enable them to get data out easily, effectively, and quickly, so you can slice and dice the way you want to see it.”
W’s system also reduces data processing time. Whereas older types of software caused an operator to come in at night to run processes taking four hours or more, now it’s all done in minutes, during the daytime.
The prospect of integrating with Excel raises another question, this one regarding how well each software package needs to work with others, including field monitoring/automation, inventory, and others. Waldroop said, “You’ve got revenue and division of interest and financial accounting and cost accounting, and those functions all really need to be well integrated, along with production and land systems. Those functions are highly dependent on each other.”
Other software, such as field automation, may not need to interface with the accounting database, but still has data that needs to be easily accessed by those who need it.
In previous years end users focused on getting the best software for each department, which could create challenges in integrating all of them with separate databases and user interfaces. Now, systems like W’s integrate many functions into one, which saves time in training, upgrading systems, and integrating databases.
Software-related efficiencies will only become more important as markets continue to fluctuate, Waldroop said. “What we’re seeing is a generally depressed market from 4-5-6 years ago. What that means is that they’re going to have to find ways to cut costs, and I think no one’s really focused on the back office until now. Now we’re seeing a lot of companies saying, ‘How can we reduce costs, how can we become more efficient in the back office, how can we work more effectively?’ The answer is, you gotta have a system that can do all that easily.”
Some producers protest that they can’t spend money for new software during difficult times. Waldroop’s response is, “Let us show you the ROI. We can show you a real return in, often, 12 months.”
The human element is also in play, with systems that, however clunky they may be, have been the water the department has been swimming in for 20 years. With the previously noted “Great Crew Change” now in progress with the pandemic, some of the human resistance to change is starting to dissipate.
Younger generations “don’t want to play with green screens or remember codes to get into screens—things that we used to do 30 years ago, they’re not interested in that,” Waldroop observed. “They want something that’s easy to use and makes sense.”
The challenges of 2020 also pushed many new customers toward W Energy Software as companies needed cloud-based systems that would accommodate employees working from home. Plus, the drop in oil prices forced many operators to shed their resistance to change, as efficiency became a necessity for survival.
On the topic of advances, Waldroop sees software’s future as becoming more predictive. “I think the key component [in machine learning] is how do we get software to become more predictive? We want it to get smarter so it can do more work for us, so we don’t have to do it ourselves, manually. I think we’ll see a lot more of that in the next couple of years.”
One example he sees is that measurement software will soon be trained to analyze outlying information. If production levels change significantly from one day or week to the next, data analysts will “teach” it how to recognize outliers and then how to perform complex analyses in order to identify the cause.
IT is definitely changing the employment landscape. While some functions require fewer people as IT moves in, the need for data scientists who can teach AI and machine learning to do complex analyses is growing. The Great Crew Change is indeed adding younger workers, but fewer are coming on board than are leaving. So the “change or die” motto extends to the ability for companies to have enough staff to get work done at all.
_____________________________________________________________________________________________________
By Paul Wiseman