Buzzword: Big Data

We all acknowledge the increasingly important role that data will play in the oil and gas industry moving forward, but overuse of the term “Big Data” has made it tempting to think of it as more of a buzzword than an industry-changing phenomenon. This may be why we’ve turned to “digitalization” and “Industry 4.0” as newer terms for the similar changes occurring throughout the modern oil field.

Having a large amount of data and needing a better way to analyze it, however, is far from new. While the preliminary steps up to the foundation of modern “Big Data” aren’t necessary to understand its evolution, gaining perspective is important. If engineers and scientists have been struggling with this problem for this long, why haven’t we been able to implement Big Data at a large scale already?

Big Data ROI. There are two major concerns: cost and availability. Though the amount of data that we can collect and store is increasing exponentially, the “cost”—that is, what companies get back from using such data—has not been increasing accordingly. The same can be said of computer processing. What we must figure out, and something I think that the industry at large has been doing much better recently, is the actual ROI for data collection and analysis.

One of my responsibilities at NOV is researching new technologies from start-up companies to understand what they do and if there are synergies between their capabilities and ours. Some events will allow these smaller companies to seek funding or acquisition from larger OFS players like NOV. Our interest should come as no surprise, especially given that at least 50% of startups are now linked to the IoT, Big Data, digitalization, or blockchain.

Actually using the data. A misconception we’ve often heard is that the purpose of data is to tell someone about a problem. The way to make data useful, however, is not only to understand the problem but to use the data to develop an actionable plan. The best actionable advice comes from combining data and data science with domain expertise—that is, making sure that the designers of a machine or system work together closely with data scientists.

Over the years, the oil and gas industry has collected an enormous amount of data, E&P companies, for example, have substantial amounts of data on their reservoirs, though historically, these data have not been used to a large degree. This trend, however, is reversing. Operators hope to take acquired reservoir data and analyze both existing and future fields, to improve efficiencies and increase resource recovery. Systems are being developed to make data collection and analysis faster and more accurate, but to truly understand the data, reservoir engineers must collaborate with data scientists.

Several key technologies will be critical to the continued growth of data use in improving operational efficiencies, including wired drill pipe, digital twins, and condition-monitoring sensor arrays. We see the use of such technologies enabling real-time data transmission, followed by rapid analysis, and ending with enhanced decision-making. Some developments are tied directly to understanding what’s going on downhole, helping E&P companies to use data to better place their wellbores and maximize recovery rates. Others will help with more effective maintenance plans, moving us from traditional calendar-based systems to condition-based maintenance plans, based solely on actual equipment use, not an arbitrary day or number. Given that some predictions are for maintenance capex budgets to be sustained in 2019, how effective companies are in adhering to those budgets will be critical.

Fracing applications. Something else that we at NOV are doing is working on using data within the hydraulic fracturing market. By using edge devices on fracturing equipment to send real-time field data to a cloud-based industrial data platform, we provide the industry with a two-fold benefit. First, we better understand, as an OEM, how to design, engineer and manufacture equipment that is more reliable and rugged for our customers. Secondly, we help those very customers gain insight into how their equipment functions, enabling them to create more effective practices. This all comes together as part of a data ecosystem that we’re trying to build, as NOV never owns the data—we merely steward it.

Perhaps a prudent course of action in discussing “Big Data” would be to replace “big” with “smart.” The implications of such a word are clear. I like “smart” better, because this indicates that we can do something with the data, not just gather it. WO