Many challenges and concepts inherent to the oil and gas (O&G) industry are equally represented in other areas of the energy sector. While the O&Gindustry has always leveraged data, it lacks the created intelligence and industry wisdom that should have resulted from the collectionof a wide range of dataduring the last few decades.Asignificant data increase is expectedas the Industrial Internet of Things (IIoTs), fiber optics, and other new supporting digital technologies are deployed throughout oilwell life cycles. The O&G industry,similar to other areas in the energy industry,has a foundationin and reliance on physics and engineering-based approaches and the simulation of complex entities. The contribution of first principles and simulations to the success of O&G operations is significant.Data-driven analytics can augment gaps in knowledge and actions to help increaseefficiency and cost-effectiveness,which are particularly importantsince the 2014 oil price decline.A fully matureanalytics value chain, asused by other energy industry peers, should be a significant asset to the O&G industry as a whole.
The full O&G industry analytics value comprises three maturity levels (Figure 1):
Figure 1. Maturity levels.
Analytics is defined as any process that creates information from data, and it is the simplest method to observedata. For example, for many workflows in the O&G industry, process logs are captured and stored in comma-separated or text files. These logs can contain thousands of lines and be reviewed in programs such as Excel. The process of reading these logs and understanding the user activities in the workflow is analytics. However, creating value using this method can be timeconsuming and human intensive.
"Data-driven analytics can augment gaps in knowledge and actions to help increase efficiency and cost-effectiveness"
Advanced analytics is the process used to understand cause-effect relationships; inthe O&G industry, it is performed after an event to understand possible causes. It also includes the creation of key performance indicators (KPIs) and a dashboard. At present, both activities are performed using limited data and in the silo withina department of an organization. For example, in a workflow analysis case, cause-effect relationshipsmight be determined in one business unit, and the impact of this unit’s activities might be discussed but not shared with other units, even though the workflow spans units.
Big data analytics is creating actionable insights in near-realtime or realtime to allow resource optimization during the oilwell life cycle. Big data analytics includes predictive, prescriptive, and cognitive models to create value from data, which allows breaking down data silos and data pipelining throughout the reservoir lifecycle, which are essential for the industry.
Analytics Maturity Level
While the O&G industry invested significant resourcesto create value from high-performance computing of data sets related to seismic studies well before many industries could exploit such computationally intensive, complex scientific data mining and visualization, it has not created similar value from drilling, completion, and other data associated with life cycle processes. The primary reason for this is the maturity level of the deployed analytics value chain.
Figure 2 shows a simplified maturity level score for O&G industry analytics across the complete value chain. It should be noted that some fundamental challenges in the O&G industry are not experienced by most industriesduring the analytics value creation life cycle. Furthermore, the O&G industry operates in a multilayered, complex, and geopolitically influenced regulatory environment, as well as posing ahigh humansafetyrisk.Hence, the value creation from data through analytics needs tobe personalized to O&G solutions.
Figure 2. Maturity score for complete analytics value chain: 5 is a high maturity score for most levels.
Some fundamental challenges experienced by the O&G industry include (a) mismatches in the frequency of data collection, which occur for various reasons, including but not limited to the calibration of meters and sensors by operators, latency in communication protocols, disconnected human interactions,such as leaving two sensors in control of two different people, and not collecting data at the same frequency, etc.; (b) the mismatch of time clocks on various data aggregators at remote locations and back offices; and (c) the data control by various stake holders with varying degrees of governance, power play tactics, and practices.
However, a path forward exists for maturing the O&G industry complete analytics value chain—Halliburton’s Smart Transform™consulting service approach (Figure 3).
Figure 3. Smart TransformSM service approach.
This solution focuses on creating value from O&Gindustry data by leveraging a comprehensive view of data to provide insightfor action throughout the reservoir well life cycle. Itincorporates best practices for data ingestion, data mining, and insights generation by managing the big data technology ecosystem and by connecting previously unconnected data sets. The solution provides for data driven models on historical dark data, the building of predictive models, and in some cases, operational cases extending those models for real-time predictive insights and action generation. As this approach matures, achieving prescriptive and holistic insights should become a reality.
As such, the O&G industry still has much progress to make because ofseveral difficult challenges related to data creation, governance, and use, but the current need for increasing efficiency and reducing cost at unprecedented levels is increasing the speed at which complete analytics value chain solutions are adopted.