Hunt Oil stays ahead of the curve with smart IoT drilling

Challenge

Hunt Oil suffered from out-of-date insights; inflexible, vendor-supplied applications; and lack of agility. It needed a faster solution for extracting information from applications to gain applicable insights.

Solution

With the help of the Spotfire® Drilling Accelerator, Hunt's Smart IoT Drilling System uses Spotfire software, in combination with streaming and messaging capabilities, to provide a single, real-time view of analyses.

Benefits

Time-  and money-saving query-the-future insights

"Using Spotfire solutions, I streamlined the process to batch-analyze wells," said Alleman. "Process optimization helped attack some of our immediate challenges." The sooner the company can get the data it needs, the faster it can react, saving time and money.

According to Senior Data Scientist Alex Molinar, data—its volume and type—has transformed the way Hunt Oil does business. Downhole, bit assembly, hydraulic, and pump data can now be analyzed with Spotfire software used to "query the future" for conditions that could lead to failure. The team can then evaluate and act before it's too late because its edge-to-core streaming model sparks immediate insights, as opposed to up to six days as before.

An example of a future query: "Based on the knowledge and location of the geological formation and the drill plan, tell me, in real-time, when conditions allow striking the surface with the drill bit, and the right force and RPM combination," said Alleman. For this query, the streaming engine combines real-time data from the rig and downhole telemetry, with the historical information and data science of the drill plan. The best RPM rate is determined through engineering algorithms that adjust the plan based on actual conditions.

A smart, highly agile IoT drilling system

In less than six weeks, Spotfire solutions enabled the engineering team to go from idea to testing of its smart IoT drilling system—a platform that rapidly deploys analytics models on real-time data.

Algorithms defined in Python, .NET, PMML, or other languages are injected directly into Spotfire streaming capabilities for evaluation. Because Spotfire code is visual and more understandable to engineers, the team can develop models they are confident will work and can respond quickly to fix them if they don't. Results are continuously pushed to Spotfire software for visualization and analysis and evolve as conditions change.