Simple models, automatically updated
ProductionCompass is a data funnel that transforms relevant production data into operational advice. Data entering the funnel comprises historical and real-time production data, well tests, and other production metadata such as operator logs. This data is then subjected to big data techniques such as squashing, while still retaining important statistical characteristics. The resulting information is used to build models using regression analysis and machine learning. These models are automatically updated in real time as new data enters the funnel, ensuring frequent calibration to the current operational situation. This approach reduces the need for advanced fluid modeling and human model calibration efforts in a daily production optimization setting.
You can shift focus from data conditioning and model calibration to applying your expertise in analyzing production-increasing opportunities and better balancing your bottlenecks. You can focus on taking actions that maximize production.
Empowering production engineers with real-time machine learning
Shedding light on the elephant in the room; uncertainty
A great advantage of our data-driven technology is that the uncertainties of measurements and models are tracked and quantified. How the field and wells respond to operational change is inherently uncertain, and ignoring this uncertainty will at best be misleading. Our advice are accompanied by e.g. confidence intervals, helping you discriminate between alternative actions and choose the best course. Quantified uncertainty can also be used to design experiments that examine unexplored operational settings with the aim of lowering uncertainty and increasing knowledge. This extends the realm of opportunities that can be discovered, and increases model accuracy. Shedding light on uncertainty enables a novel range of functionality which to a very limited degree exists in technologies and services in the oil and gas industry today.
Uncertainty estimates all the way through