Publications

The business processes of todays exploration and production (E&P) companies depend on enormous volumes of data and information. E&P professionals expend significant time and resources locating data and determining the quality, history, and context in which data and information were acquired or created. They also expend a significant amount of time trying to optimize the quality of their data to provide “seed data to jump-start new projects. This can be quite challenging because they may often find multiple sources for the same information. They may also find that these various sources provide conflicting attribute values for the same record. For example, consider well header information. Typically, government agencies provide the most reliable information for surface locations and operator names, but third parties such as independent oil companies usually have more accurate total depth values and bottom hole information. For new wells, the model differs. In this case, scouting agencies and/or well operators usually provide the best information for all attributes of the well header.So how do you address this issue and create the most accurate set of attributes for a record? One solution is to provide the user with all opinions from each source, as a complete record. But, this solution forces the user to decide which record is correct, and it does not resolve the issue of some attributes in a record being correct while other attributes are incomplete or incorrect. Additionally it may lead to users selecting different sources for the same data type.Solving this problem calls for a way to create a single record comprising the best attributes from all sources. The user can then use this single “blended record as a trusted version of the data for seeding new projects.