--> The Rules of Subsurface Analytics

AAPG Middle East Region GTW, Digital Subsurface Transformation

Datapages, Inc.Print this page

The Rules of Subsurface Analytics

Abstract

Oil and Gas companies are increasingly embracing big data analytics and data driven approaches in the drive to optimize development and production costs, increase recovery factors, and ultimately better understand and quantify uncertainties in their workflows. Most companies now have a digitalization program in place and are taking steps towards this data-driven future. From the projects that Teradata has conducted in the Oil and Gas industry, we believe that implementing a successful analytics program in subsurface involves following a few key rules. Firstly, they require bringing together the right people. Ideally what we refer to as “T-shaped” people – people with deep knowledge in one or more areas, but wide (if shallow) knowledge of the whole process, and who are open to trying new approaches. Secondly, the right data platform. Subsurface data certainly meets the Big Data definition of volume, velocity, variety, and veracity. Performing analytics on deep and wide datasets requires thinking about parallelism and performance – while also thinking about storage costs. Ensuring that analytics projects provide measurable business value requires us to take an agile approach to project management, and to repeatedly check the business alignment to ensure that the analytical results we are delivering are in some way actionable. Companies do not make or save money by running analytics projects – that only happens when they can take the learnings from the analytics projects and put them to use. In analytics projects, a vast proportion of time is spent on locating and preparing data. The required data may be available only in application databases, only as original files, or spread around various systems. We take an approach we refer to as “good enough data management” when building an analytical data platform, where structure and quality are applied in a just-in-time manner to meet the needs of the analytics. We will illustrate these key rules using case studies and anecdotes from past projects in Norway, UK, US and South East Asia.