Over 70 delegates attended CDA and ECIM’s joint workshop on the application of modern data science and analytical techniques to sub-surface data, as demonstrated through CDA’s first Unstructured Data Challenge.
Delegates heard from nine challenge participants, each sharing the results of their work on the same 3.5Tb of sub-surface data made available to them (in confidence) over the summer of 2016.
The results were remarkable, beginning with insights into how to deal with the issues encountered in processing the many types and formats of data and documents acquired over 50 years of UKCS exploration activity, and moving on to show real examples of the value that can be unlocked once data and documents are analytics-ready.
Participants presented on a wide variety of data science techniques, including how machine learning can be used to address fundamental data management issues in a scalable manner, how linguistic analysis alone may be used to identify formation analogues, and how, if a number of these techniques are used together in concert, a prototype end-to-end workflow may be realised for the automated creation of petrophysical models.
For more information on the Unstructured Data Challenge, and the outcomes presented in this workshop, please contact Daniel Brown.