Our Data Aggregator enables a “everything is information” paradigm, which leads to a much higher level of data utilization. So even before introducing data analytics such as machine learning and optimization algorithms, ProductionCompass ensures that the constant stream of refined and processed data holds value on its own. This is achieved by automatically washing, filtering and aggregating the data and enhancing it with statistical information to enable predictive features.

The aggregated data can be post-processed through machine learning algorithms and calibration of data-driven models. This opens up for a broad spectrum of production enhancing applications and use-cases, of which 3 are presented here.


Assessment of production settings gives the engineer quick situational awareness by providing an overview of what actually changed between two different production settings. This makes it possible to reverse adverse actions faster or build confidence in keeping changes that actually lead to higher production.

Further, the system can continuously monitor the quality of current production settings and report on suboptimal performance. To exemplify, underperforming wells may be flagged.


Many petroleum fields have wells that produce without a dedicated test separator available. In this situation, common practice for testing is to shut the well off and compare production rates before and after, also called a “deduction test”. Using statistical uncertainty information, our Deduction test optimizer automates live analysis of such tests giving constant updates on test progress. This facilitates reduced test times, which are directly linked to reduction of production losses.


ProductionCompass provides engineers with a live feed of automatically generated production enhancing opportunities, that would otherwise have gone unnoticed. The system generates explicit advice based on predictive analytics on how valves, pumps or pressures can be changed to utilize production system bottlenecks better, and thereby increase production throughput.