ANSA’s Managing Director, Alan Walsh and Data Analytics & Software Engineering Director, Maurillio Addario recently attended an Energy in Data event in Austin, TX. This year’s event was titled Unleashing the Power of Digital for Oil & Gas and New Energies. It was an immersive three-in-one event for experiencing the impact of Big and Unstructured Data on the energy industry including CCUS, gas energy storage, geothermal, wind, solar, and oil and gas. Themes and topics were focused on the integration of geoscience, data science and engineering, and endorsed by the three principal professional societies in the field (AAPG, SEG and SPE).
With panel discussions and break out sessions to attend, Alan and Maurillio had to ‘divide and conquer’ during their time at the conference. Here are top three take aways from each of them:
To get this right data engineers, scientists and business leaders need to work harmoniously to provide scalable solutions. There are so many (AI / ML) proof of concepts released but there is a distinct lack of solutions within the market.
Humans and AI need to work symbiotically. We (humans) are as good as AI, AI just does it faster. Our products need to be repeatable, not bespoke to ensure change is minimised as much as possible.
Shareholders know that oil and gas is now more finite than ever before (with the energy transition). However science is now suffering due to the ROI (ASAP) attitude by investors and shareholders. However the culture of the operator is now changing from a “we’ll call you” attitude to one of collaboration and help from all.
Role of SMEs
SMEs’ involvement in ML/AI projects is fundamental for success, but so is the understanding of what business case we are trying to address. Without it, it is very likely all you will achieve is a proof of concept.
Predictions will never be perfect but having humans receiving every decision is also not the solution. Humans have their own biases and the review process will vary greatly, even when the same person does it over a period of time. The preferred outcome has to accommodate the human bias and variability.
The integration of unstructured data, data clean up and classification using industry specific taxonomy form the basis upon which success can be achieved. The industry is notoriously bad in the way data is held, with a huge proportion of the ML/AI project efforts relating to get the data into a useful format.
At ANSA we deliver a comprehensive portfolio of data processing, analysis and interpretation solutions for well integrity evaluation, production logging and reservoir surveillance. We have been at the forefront of this dynamic sector for over 30 years and have a unique perspective on digital oilfield topology. If you want to find out more about our team of experts or how they can support your next project, get in touch.