Highlights from the Invest with James West podcast with Dr. Troy Ruths, Founder and CEO of Petro.ai, and Dr. Mark Zoback, Stanford University Professor of Geophysics, Director of the Stanford Natural Gas Initiative, and Technical Advisor at Petro.ai
James West: What is geomechanics and why does it apply to the development of unconventional reservoirs?
Mark Zoback: To me, it’s the integration of the physical properties of the geologic formation, the fractures, and the other attributes that they have in a geomechanical sense. But, most importantly, it’s the forces that are acting in the rocks. That’s at the core of development of unconventional resources, because the key technology is horizontal drilling and multistage hydraulic fracturing.
Hydraulic fractures follow the stresses in the rock. In other words, once you know what the forces are, you can be predictive about what the hydraulic fractures would like to do.
JW: At this stage, how is machine learning and AI being applied?
Troy Ruths: In a previous podcast, I talked about AI being at the bottom of the pyramid. In this respect, what we’re trying to do is make it easier for the end user to get access to data types, run the interpretation, and then collaborate and predict on those within the context of geomechanics.
The other big area is fingerprinting patterns. Let’s say you have a productivity pattern or that one interval is more productive than the other. If we can tie that to a key variable going through a nice technical analysis from the perspective of geomechanics, then you can go look for that fingerprint elsewhere.
We’ve had a lot of success doing that and I think that’s two big places where AI and Machine Learning have really helped act as a catalyst for a lot of these principles that Mark has put together in the book that he just put out.
MZ: Even after a couple of hundred thousand wells, which means a couple of millions of hydraulic fractures, we still have recovery factors of only 25% for gas and less than 10% for oil. So, we’re leaving more than 90% of the oil behind.
We can take these new ideas and then using the tools of Petro.ai, test those ideas against existing data and frame the problem in a whole new way to gain understanding. It’s one thing to have an idea, but it’s another thing to know whether that idea is going to work before you try to implement that idea at scale.
That combination of bringing new ideas and then confirming the applicability of those new ideas in an area of particular interest to a particular company—that’s where we’re going to really leverage these new ideas. We can figure out what’s important and what’s not and then try to attack this recovery factor problem, because we haven’t solved that through brute force and trial and error.
All of these questions surround the idea of vertical hydraulic fracture growth. We think about hydraulic fractures growing horizontally away from the horizontal well, but they also grow vertically. That’s controlled by the forces in rock and how you do the hydraulic fracturing. Well, if hydraulic fractures are growing up, they’re not growing out. The issues of well spacing, infill drilling, and stacked pay are all linked in a three-dimensional way to a condition presented to you by the Earth. You can’t change that. But, if you can characterize it and link it to the completion process, you have a shot at optimizing.
JW: If we defined the 2017 to 2019 era in US onshore as what my colleague Steve Richardson eloquently pointed out as “the megapad misstep era.” What would you say were the reasons some of the companies erred and how they corrected course since then?
TR: There’s substantial interaction between parent wells and wells that are landed in different zones. So, the assumption that you can take a type curve and multiply it by well counts has been proven not to be a viable way to understand how you’re exploiting that cube.
When you step into that 3D problem, you need to take into account the vertical propagation of fractures and the interactions of wells that are brought online at the same time versus brought online at different times.
All of that is explainable through these concepts that Mark is talking about and is something that you can measure and actually infer ahead of time.
As we step into this next era of megapads, people are realizing that they just can’t develop intervals. They need to develop the entire pad together.
JW:So what inning do you think the North American unconventional oil and gas industry is in now in terms of drilling and completion efficiencies?
MZ: I’d say we’re in the sixth or seventh inning with respect to drilling efficiencies. It’s remarkable what’s being done out there in terms of drilling and completions efficiency, but I think we’re in the second inning when it comes to understanding about what should be done. We know how to do it. We know how to do it efficiently, but I don’t think we really understand. We have a lot of data under our belt now, but there are literally millions of wells that could be used to exploit unconventional hydrocarbons in the Lower 48. Before taking advantage of any of these opportunities, we have to start incorporating a better understanding of what to do regardless of how efficiently it could be done.
JW: How do you and the geomechanics team bring it all together at Petro.ai?
TR: You know James, we’re trying to reinvent the workflow and I think we’ve talked about this in a lot of different ways. When I put the Petro.ai team together with Mark, I wanted to provide a team and vision that our clients could get behind and really help us reinvent the workflow. It’s going to be working with our clients to understand their challenges and their assets, but also bringing a lot of these new concepts to the table.
I really think that we’re one of the few players in the space that can bring this level of insightfulness and technical expertise, while at the same time, leveraging those millions of data points that a company is sitting on.
True to its unconventional designation, shale development requires new ways of working: new operations, new well designs, and even new science. While we haven’t discovered any new physics, the geomechanics of unconventional reservoirs has been largely overlooked in the realms of geoscience.
As a data scientist, I’ve been part of analyzing well spacing for several years – combining a multitude of factors across disciplines. It wasn’t until I started working with Dr. Mark Zoback that I realized we were approaching the problem without the most important ingredient: Geomechanical Stress.
In this post, I’ll explain how vertical geomechanical stress profiles can be extracted from ISIP measurements and used throughout an asset to optimize well spacing. This is a perfect activity for engineers and geoscientists while the rig count is down and the organization has time to update its plan. At Petro.ai, we have built a new tool that facilitates fast and accurate ISIP measurement.
Advances in Geomechanics
Dr. Zoback spent his early career measuring and characterizing the state of stress in the earth, which he applied successfully to wellbore stability problems all around the globe. Prior to tight reservoirs, breaking rock largely fell in the lap of drillers, with very minimal productive hydraulic fracturing. Interestingly, Dr. Zoback’s research was used to prevent a well from hydrofrac’ing while drilling.
Because of Dr. Zoback’s pioneering work in measuring and applying the state of stress, our industry has been able to drill more complex, deeper wells through a variety of formations and stress regimes. These techniques are now canon in the drilling doctrine. However, in the development of a shale asset, in which we fracture the entire length of the contact within the pay zone, we did not apply the same principles. As a result, we’ve assumed that when it came to frac’ing, bigger was better.
The Problem of Well Spacing
As it turns out, bigger isn’t better. Continuing to expand development has ushered in the problem of well spacing – how many wells, and how closely must they be spaced, to effectively deplete a shale reservoir? “Cube development” only increases the stakes; betting more dollars on upfront well spacing assumptions. While an operator will avoid the complicating factor of depletion, with all the Capex chips on red, so to speak. As some gamblers may know, in the long run, the house always wins. The same has proven true at the beginning of the second decade of shale development, the gamblers aren’t winning. Why not? To me, it comes down to fundamentals – the same issues that I saw as a data scientist – we are missing a key ingredient: Geomechanical Stress.
In order to understand well spacing, we need to understand the state of stress surrounding a well and the interactions created while stimulating and draining a volume of reservoir. In Dr. Zoback’s research, he does a fantastic job of blending theory, simulation, and empirical evidence to understand phenomena, leveraging all three. Dr. Zoback is able to identify the pattern, characterize it with key drivers, and connect those key drivers to the observations. He outlines and delivers an entire course on these key drivers, has published a textbook on the subject (Unconventional Reservoir Geomechanics), and collaborates with Petro.ai to create new geomechanics software tools (Dr. Zoback is our technical advisor on Geomechanics).
A very common problem is the lack of good data capture and interpretation in shale. I see lots of companies collect and store huge volumes of data, but these companies don’t take the time to interpret it. We may have an abundance of data, but most of it is bad: poorly organized and inaccessible. Further compounding the problem are engineers who are unable to quality control and make interpretations on collected data. As a result, engineers select from small volumes of good data, leading to an abundance of sampling bias in an industry that is overrun with data. My personal goal is to help customers use all of their collected data that holds great information but needs to be emancipated (I call this “dark data”).
Applying Geomechanics Understanding
I’ve had the great pleasure to work with Dr. Zoback for over a year, learning with him as we’ve tackled new and exciting use cases for our clients. I’m on the data and AI side, taking his concepts and scaling them to the level of operations a shale client requires, including handling complex development histories. The impact his research will have on this industry will be profound – it will be a central tenet for shale development.
Like most things in the physical world, hydraulic fractures want to open in the easiest direction. Stress is measured in pressure and there are three principal stresses that need to be accounted for in the reservoir:
the minimum stress (Shmin)
the maximum stress (SHmax) and
the vertical stress (SV).
The magnitude relationship between these stresses dictates the stress regime: normal, strike-slip, or reverse faulting. We can discern the vertical stress from the weight of the rock column, it’s hard to know SHmax, and we can measure Shmin (in most cases). In each of these regimes, the plane of the fault will be different because the fractures are opening in the direction with the least stresses.
Dr. Zoback and Dr. Lund Snee recently released a new publication that maps the orientation of SHmax and relative stress magnitudes across North America. Because SHMax is hard to interpret, they’ve done the hard work for us. Now, with their data set, if you measure Shmin (which we will explain later), you’ll be able to determine all three principal stresses in your asset.
We put this stress map in Petro.ai so you can easily reference this information across your asset. Understanding these principal stresses can have a dramatic affect when optimized: controlling for all other factors, wells drilled in the “correct direction” – 90 degrees from SHmax – perform 10-30% better.
When the pressure in the wellbore is higher than the minimum stress, it is easier for the fluid to fracture the rock and enter the reservoir as a frac than to stay in the wellbore. The wellbore pressure (measured as treating pressure at the surface) needs to overcome pressure loss over the perforations, cement issues along the wellbore, and stress shadow effects from neighboring fractures. The stress shadow effects can artificially raise the least principal stress, forcing a screen out or stopping fracture propagation.
The same logic applies vertically – whether or not you have stacked pay. In order to determine if the frac will stay in zone, you need to know the relative magnitudes of the stresses above and below the well. If the stress is lower above the well, the frac will go up; if the stress is lower below the well, the frac will go down. Many operators seem to assume that there are frac barriers (higher least principal stress) above and below a pay zone – this is very rare.
More likely, and in the most catastrophic scenario, you have an elevated pore pressure in your reservoir, increasing the productivity of the wells, but also causing hydraulic fractures to go both up and down (higher pore pressure increases Shmin). And, if you used the “bigger is better” strategy I described earlier, there are likely to be depletion affects across the whole pay zone. As infill wells are placed above and below, they will compete for shared resources with the original frac and you’ve overcapitalized your pad.
Using ISIPs to Improve Well Spacing
There are many factors that could lead to changes in the profile of minimum stress. Pore pressure, stress relaxation and depletion are your most common factors. In order to understand what is driving changes in least principal stress, you need to measure it over space and time. The most abundant (albeit noisy) data source is in 1-second frac van data.
At the end of a stage treatment, the treating pressure drops as the pumps turn off and the fractures close. There is a point called the Initial Shut-in Pressure “ISIP” that is commonly picked as part of the post-stage diagnostics by the pressure pumper. Due to operational considerations, high treating pressures, and lack of consistent theory (people still argue whether you should pick ISIP or fracture closure), ISIPs are rarely picked correctly and become a cloud of meaningless data.
At Petro.ai, we took time to develop a robust methodology for picking ISIPs after looking through thousands of stages by applying reasonable physical limitations of the pressure system. First, we account for friction loss across perforations, and second, we co-visualize the ISIP in reservoir conditions. Perforations present substantial friction – on the order of several thousand PSI – driven mostly by its radius.
As pumps shut off, the effect of this friction is removed over a very short time period. By adjusting treating pressures to “reservoir contact” we can gain a better picture of true net pressures (typically no more than 1,000 psi) and more realistic ISIPs. By visualizing the ISIP pick (with an uncertainty range) in the reservoir, we can instantly QC data to ensure it falls within a reasonable pressure gradient window (i.e. below the vertical stress and above hydrostatic).
Whenever I hear a task should be automated, I hear an opportunity for it to be collaborated. You need to have a deeper discussion about your asset including ISIPs, frac gradients and your vertical stress profile. This is why Petro.ai has social comments and tags built into the interpretation process. Any “#tagged” data can easily be searched for and filtered upon.
As part of the social iteration engine, engineers and scientists can create different model scenarios to compare and contrast their ISIP interpretations and collaborate to develop more comprehensive interpretations. As Dr. Zoback says “just because you have a solution, it doesn’t mean you have understanding”.
with David Cannon, Senior Vice President of Geoscience and Technology at Diamondback Energy
The views expressed here are not the official views of Diamondback Energy, but those solely of the subject of this interview.
If you missed the first part in this series, you can find the start of our conversation here.
What do you see as some of the key technologies that are going to help you succeed in this price environment?
There’s a lot of very cost-intensive technologies out there. During good times in the oil and gas industry, people go out there and spend millions and millions of dollars collecting high-resolution data: microseismic surveys, wide azimuth seismic shoots, pilot holes, logs in the laterals, and so on. This data comes at an extremely high cost and can give you very detailed data from their collection, but during times such as these, they’re usually the first items cut, because they’re considered discretionary spending. To follow the theme from my first post, the social proof is no longer there. So during poor price environments, we must fall back upon our scientific understand and dust of our mental models to contextualize all that high-end data we collected during the good times.
Pilot projects, high-density data pad projects, a lot of those wellbore centric data collection efforts are really just that—wellbore centric. What does that tell us about a complex reservoir 2 miles away? It doesn’t really tell us much. It’s hard to take that high-resolution data in a very finite area and extrapolate that to other parts of the reservoir without knowing context.
Geologists utilize many mental models. That’s what we do day in, day out. Sequence stratigraphy is a mental model. Structural geology concepts are mental models. Depositional environments are mental models, and we have to take all those data sources that we’ve collected and put them within that mental model. This is a key step in our progression toward data analytics and machine learning.
In data analytics and machine learning, the algorithms need context. You can’t just throw numbers into a black box and expect results to appear that make any physical sense, so you need to contextualize these things, that is where I’ve seen successes recently. Data science and machine learning are getting better with the contextualization. That has been my biggest issue with data science in the past. Very early on, it was very black boxy. You threw your data in the box, never to be seen again, then a result came out. Well, how did that work?
At Diamondback, we’re very curious about that workflow and critically interrogate whoever we partner with that provides these services. We want to see how it works. We want to be able to understand the entire process from data loading to result creation, because if we can’t explain it physically, then there’s no point acquiring it. It’s just a number, it’s just a result at that point, completely unknown with no uncertainty bounds whatsoever. How can I put any trust in that value if I don’t understand it? That’s where a lot of advancements have been had, and I think will continue to be had, in 2020 and beyond within this particular industry—pushing geologic contextualization within data science constructs.
What successes have you seen with data science or machine learning?
At Diamondback in particular, I think we’ve run the gamut. We’ve done some projects around machine learning early on, back in 2016. One of the questions we wanted to answer was:
Here’s a whole mess of data on horizontal wells—gamma ray profiles, stimulation designs, post job reports, drilling data. Here’s all the geological information, what formation they’re in, all this other stuff. And basically, they smooshed it all together and they spit out a couple of answers for what drives production. Those two things? Better production werefrom wells that were toe up and had little undulation. So does the subsurface play a role at all in this? And when we were interrogating the provider on that, they said, “we don’t have enough data”.
What does that mean? We gave them 300+ wells to utilize. But it turns out that’s not nearly enough data. We needed an order of magnitude more. In that particular example, what really enlightened us was the effect of data population. In order for an algorithm to properly learn through data, you need a ton of it.
If you don’t have a lot of data, the learning band that algorithm has is very, very narrow. You do not create an environment for that algorithm to adapt to new inputs, to outlier inputs. If you have a population of 200 wells and you have knowledge that 20 of them have outlier type production, that’s 10% of the population. Those outlier results start becoming part of the distribution. That’s a problem.
One of the things we see as a success with our partnership with Petro.ai is proper contextualization of a data science project. Tapping into geoscience expertise to build better contextualization, in addition with more data, will result in better outputs. Outputs that we can trust in our business decisions of directing a multi-billion dollar capital budget.
It seems like Diamondback has had success with data trades. Do you see that as a trend in the industry or is that sort of unique to what you guys are doing?
No one outside of the majors has the type or volume of data necessary to make a machine learning algorithm output results that are meaningful. The majors have thousands and thousands of wells, which is why they have whole departments dedicated to data science and machine learning for their own internal use. There’s a reason why, a lot of times, they don’t contract vendors because they have their own data scientists building the algorithms themselves and using them within the mental models at Exxon or Chevron.
For people like us, we have to rely upon outside data sources, either through data trades or incorporating other means of information, and that’s one of the things we’ve been able to do very successfully, acting upon those data trades and squeeze a ton of value out of that data.
One of the justifications I’ve used with management teams for high-resolution data is how this expense can be scaled to how many trades I can make with the data. If it costs $1 million dollars to collect and analyze a core, it will theoretically cost $167,000 because I will trade this data for 5 others cores nearby with competitors. That in turn will allow us to gain more knowledge on our reservoirs without increasing capital cost.
For our industry to continue to move forward as the preeminent source of energy, either stand alone or in conjunction with other sustainable energy sources, we all have to work together. Working with others with different perspectives keeps our innovative skills sharp. If you are insular, you fall into complacency.
We have built a system that’s competitive through public trading of companies. We want to compete with Apache, Concho, and Pioneer, but the problem is, we’re all extracting the same commodity. It’s not like my oil is any different than Pioneer’s oil and it’s not like the end game for that particular barrel of oil is any different either. It’s still going to get refined into the same suite of products that we all use and enjoy in our modern society.
We need to start combining our efforts to focus on more industry-friendly consortiums that allow us to share not just data, but share ideas. That’s really what it comes down to at the end of the day. If your rock is good, your production is good, full stop, but more efficient ways of extracting could potentially lift areas that maybe have marginal rock to make it economic. That is the power collaboration. We should all be working together to continue our ability to provide energy to our world to lift up the human experience.
Are there any good books or blogs you would recommend?
I’m a huge fan of social psychology; or the study of human interaction. Also, I am a fan of how human interaction and environmental interaction creates the basis for all cultures and social constructs. What are the things that you can that you can learn from interacting with humans that can provide a better basis of doing business with each other? There are a couple authors that I really enjoy reading books from. One is Yuval Noah Harari. He wrote one called Sapiens: A Brief History of Humankind. It’s a real interesting read about how our species evolved and got to the ‘top of the pyramid’. He’s written a few others. One of them is Homo Deus, which is about how our technological adaptation and our ability to communicate on a much grander scale now because of the proliferation of information. This has set into process an evolutionary change within humans that could be permanent. And as we move forward, he believes, this is the time where we may be able to classify ourselves as a species as something different from Homo sapiens in terms of our ability to coordinate with each other and how modern stimuli have changed us mentally and physically.
The other one I’m reading right now is 21 lessons for the 21st century. It’s a very good book. I’m about halfway through it right now. Lot of interesting topical things in there that contextualize modern issues. The other author I really like is Malcolm Gladwell. He’s a journalist by profession, but he’s a fan of investigating how people interact and push ideas through society. He’s a firm believer of outliers, usually the very few within our society that push a majority of our advancements and ideas through society.
How would you describe Petro.ai?
Petro.ai provides a differential product that allows our company to have an edge on understanding what we’re doing in the subsurface and how we interact with it.
As we were selling this project internally, I think what differentiates you from other people, is the fact that, simply put, you provide the platform and access to expertise to allow us to thrive within your platform. And that is something that is differential. There’s a lot of people who push machine learning, who push their own platforms for being able to come to terms with how we interact with the subsurface, but it’s usually all project-based. They don’t give you the keys. It’s like an Uber. You have to call them and they show up and they pick you up, whereas you guys are selling us the car. We’re able to drive it ourselves, but then we go back to the dealership every so often: something’s wrong here and you help us fix it, but we still own the car at the end of the day. So that’s kind of how I view Petro.ai.
That’s the kind of partnership we want, as I said before, we want to understand. We want to be able to know where to fix the issue if something breaks down. That is important to us because it allows our organization to learn and adapt in the proper manner to new information.
Petro.ai uses AI and Machine Learning to empower domain experts, data scientists, and executives with instant information in the right context, energizing teams and transforming data into action. Partnerships with global leaders in cloud computing (AWS) and geoscience (Dr. Mark Zoback) allow Petro.ai to deliver differentiated technology and teams to Accelerate Discovery.
with David Cannon, Senior Vice President of Geoscience and Technology at Diamondback Energy
The views expressed here are not the official views of Diamondback Energy, but those solely of the subject of this interview.
Tell us about your background and what you do now.
I’m the Senior Vice President of Geoscience and Technology at Diamondback. Mostly my purview is on the subsurface side—being able to better characterize and extract Diamondback’s resources in a cost-efficient manner.
I started out in the industry back in 2007, a geoscientist by trade and by formal education. I got my Master’s Degree at Penn State University in Structural Geology with a focus mostly on rock mechanics and fracture mechanics, so I’m one of those weird geologists that actually likes math.
In our industry history, geoscientists have always worn the mantle as being those who spend all the money without regard for economics, under the guise of pursuit for truth. As I moved into the oil and gas industry, I think I’ve always had an inherent businessmen’s mind, which resulted in marrying science and technology of the subsurface into sound business decisions.
I know there’s a lot of others that think this way, too. Can we marry the pursuit for data and truth in the subsurface with sound economic practice? We can do these things in a cost-efficient manner that not only allows us to succeed in terms of creating a business case for data collection or data manipulation— which is where I see players like Petro.ai being an integral part of—but that also allows us to get stakeholders in the process.
Engineers tend to have an economic mindset: what’s the most cost-effective practice to be able to maximize my return? And if you can sing along those same lines, I think that you can get a lot of buy-in on new technologies, a lot of buy-in on something that might be disruptive for your particular industry. That’s honestly the way all companies should approach it.
We have to stop thinking that this is an antagonistic setting where the geologists have one way of thinking, engineers another, and the finance guys another. We all have to meld our thought processes together to be able to get the best outcomes.
Why do you have a passion for change in this industry?
Our industry should always have a passion for change. The way I see it is we must always evolve, and if we do not evolve, our industry will perish. It’s borne out in the data. You see historical references to it all the time.
One particular industry that has gone through a period of low innovation, and now feels the repercussions, is the coal industry; one of the preeminent sources of energy for the world from the 1800s through I think about the 1980s. Well, they got complacent. That complacency led to a slowdown in innovation around things like more efficient extraction techniques and more efficient, environmentally friendly conversion techniques. Are there ways to thermally alter coal in a way that can reduce the amount of greenhouse gases and reduce the amount of pollution that is released into our environment? I just don’t think they were ever ahead on that respect.
They always reacted whenever new regulations came out or whenever there was social pressure. Only then would the coal industry react, and they usually reacted in a very minimal way, just enough to get by. They were on top. Coal would reign forever. Why would they have to change?
Technology evolves as new social pressures and new paradigm shifts occur throughout our human society. Other alternative sources for that energy started to come to the forefront, and I think the biggest displacement technology we see now is just switching over to natural gas. It’s simple, right?
You can take a coal plant and convert it to a natural gas plant relatively easily, and with the advent of horizontal drilling and hydraulic stimulation in our industry, this resulted in an oversupply of natural gas. That fuel source became extremely cheap, so all the utilities that ran these coal fire plants saw an opportunity to behave in a much more efficient manner. At the same time, gas was able to better answer the calls for change from environmentalists and from general social pressure around pollution. So, they made the switch.
More and more coal plants are continuing to switch to natural gas even today, and the market share of coal continues to slide. They are no longer the number one producer of electricity in the world. They’ve been displaced and they will continue to be displaced. The only reason why they were displaced isn’t really because natural gas is just better. It’s because they stopped innovating. They stopped seeing where they fit within the future and seeing how they can adapt to that new reality. And that is a problem that a lot of industries have, and that’s something I’m passionate about with our industry, as I can see us traveling down that same road really easily.
If we continue to be complacent about where we fall within the energy industry and overall giving energy to the human population, we can easily be displaced if we’re not thinking about new technologies.
There’s some research that’s being done now with the Earth and Mineral Sciences Department at Penn State in conjunction with private industry in Appalachia, looking at ways to take natural gas methane streams and thermally alter them with microwave plasma. The process breaks down methane and converts it into molecular hydrogen, for hydrogen fuel cell technologies, and graphene for structural additives to steel and concrete. So, they’re thinking about how our industry, the hydrocarbon industry, can be a part of the solution of renewable energy.
It’s not an or statement. It’s an and statement. And that’s one of the things that I’m passionate about is that our industry can be a part of that solution. We just have hurdles, and it’s mostly philosophical hurdles, of people saying, “we’ve always done it this way. We’ve always provided energy this way. This is how we’re going to do it.”
That is the mentality that kills you. That’s exactly what happened to the coal industry in the 60s and the 70s, and they refused to change because of that mindset. They basically said, “where else will you get your energy?” Well, they found out where else the world could get their energy. So, we can’t stand by and let that happen. It will be the death knell of our industry if we don’t find ways to couple into the new research around providing energy in a more sustainable way to the world.
It’s a full cycle problem. It’s not just extraction. Extraction is one part of the problem. It’s also taking that product, then converting it to the energy that’s consumable for whoever your consumer base is. That’s one of the things that I think we have an issue with is that we’re always really decoupled in that.
The independent E&P companies, they’re just worried about extraction and production, right? Once they sell the product, you’ve ended the value proposition. End of story. Move on to the next barrel. But some of the engineering, some of the science, some of the application of technologies that we’re using in the extraction realm could also help on the consumption realm, because at the end of the day, the physics, chemistry, biology that we use on a daily basis wraps up into the work done to convert that resource to energy.
Are we going to completely focus on the science of resource conversion? No, because we have our business models that state that we have to extract this resource. But are there ways that we can bring ideas to the forefront that we could change the game on how those resources are then used? Yes, and we as extraction companies should be part of that research and ultimately that solution.
In 2015 and 2016, the downturn gave rise to this digital transformation in oil and gas. How do you see this 2020 downturn affecting things?
I always fall back upon is the diffusion of innovation. It’s a concept that was actually written back in the early 60s talking about how technology or even an idea, moves through and is adopted by a population. It’s an elegant construct really.
When a particular idea or technology starts, you have the innovators, the ones who are creating that process, the ones who are getting the bloody nose because they’re the first ones through the wall. Then you have early adopters, folks who say, “That’s a really cool idea. I want to adopt it and potentially make it better. Let’s push this technology forward.”
One perfect example is Elon Musk. Plug-in electric vehicles have been around since the 1960s. They actually made functional models of plug-in electric vehicles back then, so what he’s been able to deliver with the Tesla vehicle is nothing new. He was an early adopter of that technology. We have better complementary technology around battery tech and energy efficiency to increase range and operate more technologically advanced vehicles. So, he’s an early adopter. I wouldn’t consider him an innovator.
Then once you go past the early adopter phase, you have what’s called the technological chasm. It’s at that point where the technology has to reach some sort of social proof, and social proof can be defined by anything, depending on the social system that you’re trying to push an innovation or an idea through. That social proof can change and vary, dependent upon the answer and solution that technology is trying to address.
To continue with the example of electric vehicles, it’s usually cost and value. A middle-class family of four in Iowa wants not only a vehicle that has a long-extended range, which Tesla has, but they also want a vehicle that has a low cost. When that family buys a Toyota Camry for $30,000, they get the benefits of long range and value. This is why the Tesla Model S hasn’t jumped the chasm, because that vast consumer base, middle-class America, cannot couple range AND value. That’s why electric vehicles have not jumped the chasm, because right now they have not attained the social proof of value.
In our industry, oil and gas, the social proof for technological innovation is also monetary. It’s essentially the price of oil. If the price of oil drops below a certain threshold level a lot of technological advancements can’t jump the chasm to get more adaptation through our industry.
What we do, what we spend money on, is inherently tied to the price of the product we sell. How much money we make, how much revenue we make on a quarter over quarter basis, year over year basis, is going to be our war chest to be able to go out and spend dollars on innovation and potentially be more efficient.
During times like 2015, 2016, and also today, are times where the social proof concept falls apart for our industry, and a lot of innovations fall backwards in that curve. They no longer have social proof, and people start to drop that technology because it’s no longer viable within that social proof context.
The interesting part about that entire thesis from E.M. Rogers around technological adaptations is when you have something that falls out of social acceptance, something else replaces it. There’s always something waiting in the wings to get social proof in times like this. The constant struggle with competing technologies as social proof shifts and changes with time causes a very stilted technological history within our industry. When times are good, we focus innovation on subsurface assessment. When times are poor, we focus innovation on operational efficiency.
Things like data science and machine learning met social proof in 2015 and 2016, as they were disruptors on the efficiency side. Then as things started getting better, there was a stasis, a plateauing of data science. There wasn’t this huge rush of, “Let’s keep pushing the paradigm and pushing the technology for our industry.” Instead, we fell into, “Well, this works, let’s just keep using it.” I predict we will see another creaming event for those technologies in 2020 and beyond, because I think the learnings that were acquired in 2015 and 2016 are going to be brought back up to the forefront and the value propositions are going to be shown again. Then there’s going to be more people paying attention to the technology. In turn, that attention is then going to continue to evolve it, and that evolution then keeps that technology moving forward and doesn’t allow it to get complacent and lag like the coal industry did.
with Kyle Gorynski, Director Reservoir Characterization and Exploration at Bonanza Creek Energy
If you missed the first part in this series, you can find the start of our conversation here.
Why do you think there’s been so much hesitation around change?
There’s a strong engrained culture in oil and gas with a generation of people who’ve been doing this for decades – although on completely different rocks, play types and extraction techniques. There’s been pushback on adopting new geology or engineering software because people become so comfortable and familiar with the tools they use. I think a lot of it is simply the unique culture in oil and gas. Shale is still new, and our guesses are still evolving. I think we’re getting closer and closer to what that right answer is.
New organizations will have to adopt new technologies and adapt to new trends, because there’s no other way to make this business work.
As a scientist or engineer or manager in this space, we really have to be cognizant that the goal of a lot of vendors out there is not to help you get the right answer, it’s to make money. The onus is on us to vet everyone and make sure we’re getting the answers we want.
Machine learning is simply another tool, like a physics-based model, aimed to help us predict an outcome and increase the precision and accuracy of these predictions. People have made pretty fantastic inferences from these kinds of tools.
You can’t just pay a company to apply machine learning to a project. You need to help them utilize the correct inputs, the relationships, and ensure the predictions and outcomes match the other observations you have from other datasets.
I don’t think any organization should be cutting-edge for the sake of being cutting-edge. The goal is to solve these very specific technical challenges quicker and with more accuracy. Our job is to extract hydrocarbons in the most safe, environmentally friendly, and economic fashion. Technology like machine learning and AI are tools that can help us achieve these goals and needs to be done correctly.
Can you share any successes around data science or machine learning at your company?
The industry has been using these techniques for a long time. In their simplest form, people have been cross-plotting data since the early days of the oil and gas industry, trying to build relationships between things. At the beginning of my career, I remember using neural networks to predict log response.
Now we use predictive algorithms to help us predict log response where we don’t have certain logs. Let’s say we want to predict lithologies—carbonate-clay-quartz-feldspar content in a well— we’ll build relationships between triple-combo logs, and more sophisticated, but scarce elemental capture spectroscopy logs. We don’t have ECS logs everywhere, but we have triple-combo everywhere, so if you can build a relationship between those, then you have a massive dataset you can use to map your asset. That’s a simple way we use this type of technology.
Like almost every company now, we’re also predicting performance. That’s how we’re able to make live economic decisions. We have a tool where we can put in a bunch of geologic and engineering inputs and it’ll predict production rates through time that we can forecast, add new costs, and run economics live. We’re running Monte Carlo simulations on variable rates, volumes, lateral length, spacing, commodity pricing, and costs that are based in our best estimates to predict tens of thousands of outcomes to try to help us better understand what the best decision could possibly be. I think that’s the most impactful place it’s being used, and I think that trend is being adopted more and more in industry as I talk to my peers.
Type curve generation is no longer grabbing a set of wells and grouping them together and fitting a curve to it, but it’s trying to predict the infinite amount of outcomes that are between the extremes.
Have you seen any success among your competitors using technology, specifically data science and analytics tools?
There’s some great work out there across the board. I had a lot of fun at Encana (now Ovintiv) seeing a lot of my peers who are exceptionally smart really trying to adopt new technology to solve problems. I’ve seen some amazing work getting people to adopt new ideas, new thoughts, new predictions. I like going to URTeC. I think that’s a fantastic conference. I always find a number of great sets of technical work that has come out.
I think the industry is doing a great job. There’s a ton of really smart people out there that know how to do this work. I think a lot of young people are really adopting coding and this bigger picture approach to subsurface, where it’s not just you’re an engineer or you’re a geoscientist, you really have to understand the fluid, the pore system, the stresses, what you’re doing to it. There’s no way we can be impactful unless we understand the really big picture, and people are getting much better at that, trying to use tools and develop skillsets that allow them to solve these problems a lot quicker.
How would you describe Petro.ai?
We see you guys as filling a gap we have. It’s the ability to pull data together. It’s the ability to simply apply big data to a dataset we quite frankly don’t have the time or the capability to do in-house. Petro.ai provides us with a very important service that allows us to get to a point that would take us 12-18 months to get to on our own, but in only a couple months. What we really like about it is the fact that we’re developing something that’s unique and new and therefore has our input and involvement, so we’re not just sending you a dataset and asking for an answer, we’re trying to say what we think drives the results, and we also want your feedback. So you’re also a group of experts as well that not only have your own experiences, but you’ve seen people’s assets and plays and how everyone else in industry is looking at it, so it’s nice to have this group of consultants that have the same goal – to address a problem and try to figure it out. We want to get to an answer as quickly as we possibly can and start to apply those learnings as quickly as we possibly can.
Learn how Petro.ai merges theory with operational data to inform asset development strategies in this short video with geomechanics expert, Stanford professor, and Petro.ai Technical Advisor Dr. Mark Zoback.
Every well that gets drilled is an opportunity to gain more insight and understanding. The team behind Petro.ai believes in building tools for people to use the data they have to draw the right conclusions. How can machine learning aid geomechanics? Learn more about the Petro.ai approach in the above video.
What are the opportunities and challenges with unconventionals? What can geoscience offer to unconventional development? What role can data science play? For a brief discussion, watch this video of Petro.ai Founder and CEO Dr. Troy Ruths with Stanford University Professor of Geophysics, and Petro.ai Technical Adviser, Dr. Mark Zoback.
The next installment our Q&A series with geomechanics expert, Stanford professor, and Petro.ai Technical Advisor Dr. Mark Zoback touches on how a better understanding of geomechanics can lead to completion designs that improve well performance.