Increasing Shale Productivity Happens Best with Data (And we’re sorry about the buzzwords)
Data Science & Analytics

Increasing Shale Productivity Happens Best with Data (And we’re sorry about the buzzwords)

Rosemary Jackson  •  

Petro.ai: As a technology company we are so relevant, so advanced in the industry that we’re every 2021 buzzword and more. We know you’ve been burned, but for us those buzzwords are real. Petro.ai has real Artificial Intelligence, AI, that’s combined with the data you really need to be productive in shale.

Data Management: We fit data into highly accurate workflows. We’ve seen productivity increases using data that’s been hidden as unused dark data in the oil and gas industry. Look at what you’ve got. Transform it and use it. Your data is your way, your only way to make changes in your company, in the industry, in the science that finds itself lost in the folds of intuition and legacy concepts.

Horizontal Spacing of Wolfcamp A

The goal of Petro.ai is to embed your data into AI-derived models that accurately evaluate your tradeoffs and opportunities in your unconventional reservoirs increasing your productivity. And for that, you need to know what data to use, to have that data transformed into the Petro.ai platform, and to click on an app and bring that data in for analysis. To start that process, come to your data with questions.

The first risk for most companies is whether they’ve put their data in a place that can be accessed,” Dr. Derek Ruths, Chief Data Scientist of Petro.ai, explains, “For most data it’s not a problem of dirtiness, it’s a problem of where you’re going to put it and how you’re going to access it. Data has to be transformed. The first transformation is accessibility and some clean up.

The second risk which is also an opportunity, points to the concept of dark data, all the data that a company has collected and isn’t being engaged. And the reason it isn’t being engaged isn’t because it’s not useful, it’s because the company has put itself cognitively in a place where it’s not thinking that it’s useful. The second transformation of data is pulling out useful dark data that you already have.

“The opportunity we’ve recognized as we’ve moved into this era of data analysis and data science is that there are mineable insights from dark data.

“The opportunity for a company that shines a light on its dark data in a systematic way, is it can generate or discover insights that will improve their operations, improve efficiency, improve their productivity.

“And effectively for free because it’s in the data itself. They’re not having to pay for more data, not having to go out and look for it, they just have to curate it.

“When we think about how we realize that opportunity, there’s a couple of important steps. The first is being aware of what your dark data is. And here I’m talking about the data you keep that you don’t use. And for most companies that’s about 90% of their data. The first step is just being aware that you have information that is useful and the creative questions to understand that usefulness.

Now, you can be as creative as you want or have a great idea about the value of your dark data, but if it’s not organized in a way that you can actually use it then it’s effectively useless. Dark data can often be locked in the dark by not being cleaned, not being well maintained, not being stored in a way that people can get to. There’s all kinds of data access and data engineering questions that have to go into making dark data, unused data, useful.

For oil and gas companies, dark data can mean everything from microseismic to reservoir characterization to pressure data, including the instantaneous shut-in pressure values that are collected during each hydraulic fracture event.

Midland Basin Pad Scenario Designer

ISIP data holds a lot of diagnostic value,” Dr. Nitin Chaudhary, Senior Data Scientist, emphasizes, “You’re sitting on all these wells that you have completed and fractured in the past and you’re not utilizing that piece of information. As you drill more and more wells you have a global model, which is what Petro.ai helps you develop. You have a global stress model that gets informed. The more location information that you load in from your reservoir, the more vertical profiles of fracture gradients you have created.

The Petro.ai machine learning operations (MLOps) apps grow in accuracy because the more wells you put in the more ISIP values are picked automatically, which means more data for the platform to build a stress profile which is the most important input that goes into the predictions that we develop in the Pad Scenario Designer App. Your unused ISIP data becomes an important part of your pad design and increases your pad productivity.”

Recently HBR noted that, “too often, companies discard data because it does not have value in one interaction, yet this same data may be valuable in another context. Data-driven companies continually discern that data which is truly useful and will deliver the greatest insight and business value...they teach their teams to be experts in quantitative decision-making techniques and they leverage their employees’ deep knowledge of their business when they’re building important models. Buy technology and AI platforms and workflow engines — don’t reinvent the wheel.” Data, including mined dark data, combined with the AI engine is the force of your productivity future in shale.

(Author’s note: Future casting for 2022 what buzzwords are bugging you? I’d love to know what you think. Send me your hated buzzwords, rosemary.jackson@petro.ai)

(Author’s prediction: 2022 Machine Learning Operations joins the list)

PETRO
114 Main Street Ste. 200, Houston, TX 77002
4-5609 Avenue du Parc, Montreal QC, H2V 4S8

© 2020 Petro.ai All Rights Reserved