Knowledge technology provides energy companies with knowledgeable applications that help them make better decisions faster in all decision points. One such knowledge application (CrudeFlex) is related to supporting the decisions of purchasing certain crudes to be processed in certain refineries through properly evaluating the risks associated with processing such crudes.
In this paper, we discuss the basic concepts of knowledge modeling and how specifically CrudeFlex was developed as a knowledge application, how it works and how rifineries are leveraging it to strengthen their competitive edge and proactively evaluate and manage risks associated with the crudes.
The new generation of knowledge applications are powered by a combination of computational knowledge graphs and computational algorithms. These algorithms encode the expertise of subject-matter experts, such as process engineers and combine their experience with decades of historical data extracted from databases, documents, and sensors in addition to ever-growing corpus of technical research to support better decisions faster. This technology enriches and combines companies’ internal siloed data with public data to create an integrated digital knowledge layer. Engineers can evaluate and manage the risks associated with known processing and new crudes in any of their refineries.
Refining engineers have easy access to knowledge related to people, equipment, vendors, crudes and more, so that they can make better, more informed decisions faster. In this paper, we show how the application of such algorithms helps the reading of hundreds of thousands of historical reports to harvest knowledge about the risks, and store the extracted knowledge in an enterprise digital knowledge layer, saving millions of dollars by enabling experienced engineers to make significantly better decisions faster through using the harvested and captured knowledge.