Page 394 - Intelligent Digital Oil And Gas Fields

332 Intelligent Digital Oil and Gas Fields

activity. The understanding and interpretation of such monitoring still
requires a lot of effort in tying with the governing dynamics of the complex
systems.
In another interview, Anthony McDaniels, President of Rare Petro Inc.,
addressed remote operations and control. The “technology industrial
revolution” is here and is being driven from consumer electronics. Robotics
is going to be on site soon and although humans will still be in the field, they
will be fewer and fewer.
As hardware and processing become less expensive, there is a “tipping
point” to move the ‘smarts’ (analytics) to the sensors that are on the end
devices (“edge”). Even small operators are able to collect data and in the
future will have the analytics as well. Therefore, move analytics to the
“edge” and provide decisions and guidance to users—users do not have
to spend their time managing and analyzing data. Software is automatically
upgraded and distributed to devices.
The demographics of the younger generation demands mobile technol-
ogy for everything. There will be “Uber” for supply chain, well interven-
tions, tank deliveries, etc. With GPS on every truck, pumpers will be tracked
and guided for step change in efficiency. Services will be commoditized.
Is there a potential in synergies between the state-of-the-art social/
mobile high technology and O&G optimization, and how could the
E&P digital transformation maximize the “bang for the buck”? Pallav
Sarma, Chief Scientist at Tachyus (an O&G technology company), notes
that, while techniques and underlying principles of data science have been
around for decades in various disciplines such as statistics, computer sci-
ence, machine learning, probability theory, etc., it is only recently that data
science as a unifying umbrella has received significant attention and pop-
ularity. This popularization is due to an explosion of the quantity of data
collected daily by these technology companies, a significant increase in
computational resources available, access to cloud computing facilities,
and advances in data science algorithms. A recent and highly visible exam-
ple of the application of data science in technology is the 4-to-1 defeat of
the reigning GO champion by Google’s DeepMind team. GO is a board
game orders of magnitude more computationally complex than chess;
therefore, brute-force computational solutions are not viable yet for solv-
ingGO. Just recently,itwas generally thoughtthatasolution to GO wasat
least 10 years away. However, DeepMind’s approach to GO was made
possible by access to huge amounts of training data, access to Google’s very
large GPU clusters, and significant advances in deep-learning neural