In tightly spaced unconventional reservoirs, hydraulic fracture hits communicated from child (infill) wells to parent (existing) wells has become the norm. The effect of frac hits on productivity can be positive, increasing the EUR of the parent well, negative, decreasing the EUR of the parent well, or neutral, having no effect on the EUR. Kyle LaMotta, VP of Analytics, explains the Petro.ai answer to quickly determining this effect in the newly released investigative decline curve comparison in the Frac Hit app, “Petro.ai is comparing two decline curves, pre and post frac hit, and then looking at the EUR difference between those two. The area between the two curves becomes an accurate measure of the production difference.
Updates to Treating Pressure Prediction App with Nitin Chaudhary, Senior Data Scientist at Petro.ai.
Every Friday afternoon at 3:15 PM CST, our team assembles remotely in Montreal and Houston for our Weekly Company Catch-up where someone shares what they’re working on or drops a pin on a map and tells a story or we figure out who likes edge or center brownies. Recently, in one of these catch-ups, we talked about, what is Artificial Intelligence? And I wanted to share their voices with you.
Petro.ai Howgozit Dashboards provide an information foundation for a company, part of that SSOT. They are also an expansion method for further development based on a timely incorporation of customer input. Charles Connell, VP of Development, outlines the potential, “We can respond quickly to customer requests by creating or adding to a dashboard. We use dashboards for quick answers, but also as prototypes for new apps. A lot of the workflows that we’re scoping start as a dashboard. As the feedback comes in, we add more functionality and create a fully featured app.
Charles Connell, VP of Product at Petro.ai, provides updates on the latest changes to Petro.ai.
Machine Learning Operations, MLOps is the technology that will define both AI and the O&G Industry, as machine learning models are merged with ever-changing data into a collaborative, robust and agile prediction tool. Where DevOps provided a process to deploy software into a business production environment, MLOps takes the moving pieces of training ML models and a dynamic data pipeline and strategically deploys the system into the business production environment. Petro.ai uses MLOps to achieve a high level of predictive accuracy with well trained ML models and a complete understanding of data collection, data aggregation, and data deployment.