Edge processing and data integration make the business stronger
Oil & Gas is, in one sense, a world leader in the Internet of Things (IoT). Producers and developers have been gathering data from remote sensors for more than half a century to support seismic analysis, monitor operations, and predict maintenance needs. At the same time, the industry has a long way to go to realize the ultimate benefits the IoT can provide by enhancing reliability, optimizing operations and strengthening the business.
E&P is awash in machine sensor data that accumulates faster than it can be analyzed. Thousands of sensors are transmitting vast streams of data from the field, and new data streams are coming online from drones, wearable devices, and larger sensor networks.
Exploration and production operations can generate as much as 1.5 terabytes of data per day that is transmitted and stored for analysis. Geoscientists and engineers depend on machine data to understand the basic facts about reservoirs and operations, but this is just the beginning of what is possible when sensor data is combined with intelligent, timely analysis.
O&G companies face monolithic challenges from commodity price risks to pending retirements, and leaders need to optimize data analysis to maintain a competitive edge. Effective IoT implementations can support exactly the kind of improvements that E&P needs: reductions in accidents and unplanned outages, increases in productivity and recovery, and enhanced support for business agility.
Intelligence at the Edge
One widespread problem is that the data itself is inadequately managed. When machine data is transmitted directly to a data center or the cloud without real-time analysis on-site, companies miss important opportunities. The alternative is to apply edge analytics by processing machine data near the source.
Instead of simply uploading the machine data to a data center or the cloud for eventual analysis, edge processing serves up real-time intelligence in user-friendly dashboards that help operators make fast, accurate decisions. For example, sensor data can be processed with rules programmed to alert an operator to the potential for a drill bit to get stuck. This creates an opportunity to halt drilling preemptively and prevent a costly interruption. With downtime costing as much as $1 million a day, that kind of actionable intelligence is invaluable.
Edge processing can also be used to minimize data transmission, which can be costly and inconvenient at a remote site. When an edge processor filters out non-anomalous data and selects only anomalous data for uploading, it alleviates data overload issues. Less data is transmitted from remote locations, which reduces transmission costs, and data centers are less pressured to scale in response to data saturation.
Another opportunity available with machine data is to improve longer-term analysis of exploration and production processes to improve business performance. For example, portfolio management improves when management has the data analysis it needs to rate wells by production efficiency. Similarly, reservoir assessment can improve when physics data about resources is merged for analysis with non-physics data regarding assumptions, uncertainties and scenarios.
Analysis is most powerful when applied beyond operations at the leadership level. The sensors deployed in exploration and production generate structured data that can be merged with historical sensor data and other data types to take business analysis to new heights. Leaders can evaluate existing assets with greater certainty and develop a better understanding of the revenue potential of resources they might acquire.
This kind of cross-discipline analysis can transform the value of data by supporting greater understanding of operations in exploration, extraction and production. With improved visibility, leaders can gain confidence and make decisions that improve capital and operational efficiency. This is precisely the kind of strategic advantage that businesses need to succeed in an environment of high risk and increasing uncertainty.
Leadership can drive value by supporting IT initiatives, such as deploying third-party solutions that improve data standardization and enable integrated analysis of structured and unstructured datasets. There are challenges in the realm of information management and data processing. Structured machine data, in many cases, is stored in silos in formats that are unreadable to analytics applications. When a company can merge machine data together with weather reports, maintenance records, syndicated data and other unstructured data types to derive rich insights, the data has much greater value.
Given the potential for business value, IT architecture is an important leadership consideration. IoT implementations can support improvements throughout exploration, drilling and production as well as better performance from geoscientists and engineers. When IT is empowered to seek out solutions that will optimize the power of machine data, the organization wins.
The IoT may be old hat in O&G, but companies can achieve essential improvements in performance from the board room to the production platform by processing it with powerful analytic intelligence.
— By John MacKenna, freelance technology and business writer and the former Editor of Oil & Energy.