Geoanalytics and Energy, a data-driven love affair
Mass adoption of artificial intelligence is rapidly making changes across different industries, yet it has always been intertwined with oil and gas.
According to a study conducted by Ernst & Young in 2017, it was revealed that out of 75 major oil and gas companies examined, 68 percent had made investments exceeding $100 million each in data analytics over the previous two years. Additionally, almost 75 percent of these companies planned to allocate a significant portion of their capital budgets, ranging from 6 to 10 percent, towards digital technology.
In 2013, James Crawford founded Orbital Insight, a Palo Alto based company which utilises satellite, drone, balloon imagery and geolocation data, combined a proprietary combination of artificial intelligence and computer vision, to provide insights on from data gathered to businesses. Orbital Insight began by examining large-scale satellite images of oil tanks worldwide, specifically focusing on tanks with floating lids. When these tanks are filled, their lids rise, and as they are emptied, the lids sink. Inside the tanks, the sun’s position creates crescent moon-shaped shadows. By identifying and analyzing the variations in these shadows, the company can estimate the volume of oil stored in approximately 20,000 tanks under its surveillance globally. Orbital Insight mainly focused on the emerging economies, where the are often less reliable or untimely oil reserve data. The company has proposed that China has been underreporting oil reserves declared.
Another example would be how Orbital Insight used their GO platform to for supply chain monitoring, by tracking the construction of the Brunei Hengyi Refinery. They analyzed data from multiple sources such as cell phones, ship-based sensors, optical satellites, and radar satellites. By studying foot traffic, anomalies, geolocation data, and land use, they gained insights into the construction process and facility operations. They also monitored ship traffic at the refinery berths and observed changes in oil storage levels. Additionally, they detected flaring activities using Short-Wave Infrared Satellite Imagery satellites, providing information on the facility’s operational status.
Another player, Devon Energy, has a stellar Exploration and Production team leveraging data analytics throughout stages of exploration and production stages of reserve discovery, drilling, and extraction. In terms of discovering oil reserves, they utilize data analytics to analyze geological data generated by survey techniques, comparing current survey results with prior data to determine underground reserves more accurately.
During drilling operations, Devon utilizes sensors on drill bits and rigs to track the location and progress of drilling in real-time, leading to more precise and safer drilling with fewer equipment breakdowns. For extraction, Devon leverages data analytics for predictive maintenance of wells, decision-making on re-fracking non-performing wells, and monitoring the quantity of reserves across their extensive portfolio of wells. Since starting in 2012, evon has experienced a 250% increase in 90-day production volumes, which is the highest in the on-shore market in the United States. Additionally, their cost per well has decreased by 40% during the same period.
Both Devon and Orbital Insight face challenges related to data generation and cost. Devon’s challenge lies in the expensive nature of generating data during drilling operations and the limited volume of geological data. Similarly, Orbital Insight deals with the cost of acquiring satellite images from its partners to analyze, and the limited coverage of satellite imagery, which can skew the analysis.
Devon’s challenge lies in interpreting geological data to determine underground reserves and optimizing drilling and extraction processes. On the other hand, Orbital Insight faces challenges in obtaining accurate and comprehensive data from satellite imagery. They struggle with limitations in capturing indoor activities and product storage, which can impact the accuracy of their analysis. In addition, Devon had to overcome the challenge of getting buy-in from engineers and scientists to implement a data-driven approach. This involved aligning their efforts with the goal of extracting oil and gas effectively. In contrast, Orbital Insight may face organizational challenges in terms of building partnerships with satellite owners and securing access to high-quality satellite images.
Furthermore, Devon operates in a cyclical business tied to commodity prices, requiring the ability to invest across the business cycle. Orbital Insight potentially faces challenges in adapting to changes in the market and staying competitive as other companies enter the satellite data analysis space.
To summarise, both Devon and Orbital Insight encounter challenges related to data generation, technological advancements, and competition. However, Devon’s focus is primarily on geological data interpretation and optimizing oil and gas operations, while Orbital Insight’s challenges revolve around satellite data acquisition, accuracy, and maintaining a competitive edge in a rapidly evolving industry.
In the world of data-driven energy organizations, remote equipment surveillance plays a crucial role in managing equipment productivity, degradation, and stoppages. Organizations can continuously monitor equipment, receive decision support, and automate interventions when degradation is detected by utilising numerical models that capture patterns. The result? A significant boost in operational efficiency and return on investment.
But, of course, there are challenges to overcome. Companies need to find robust ways to access and handle their data, adapting to changes in IT systems. Data pipelines need to make room for for anomaly detection based on historical data and continually improve models and rules to reduce false positives and accurately identify degradation instances. Balancing the cost of missing a degradation against the cost of investigating notifications is also crucial. To effectively manage and resolve notifications, a business process management system should be implemented for issue logging and case resolution.
By integrating data access, wrangling, analytics, and intervention management, organizations can uncover insights and take appropriate actions. This powerful combination of data and targeted analytic intelligence can lead to significant savings in operational costs. With the energy industry becoming increasingly digital, the possibilities for implementing these strategies are plentiful.
