In 2019, over a series of AAPG Analytics / Machine Learning Workshops across different regions, I posed this question to participants, “would you be ready to fly on 737 Max when it re-enters service?” In sessions attended by mostly engineers and data scientists, fewer than three hands were raised. However, in one hosted by the Kansas Geological Survey, half a dozen participants raised theirs. Interpretation? It may be attributable to professional mindsets, e.g., geologists more explorationist (risk takers) and engineers more measured (risk averse). More data will yield a better example.

Data example? It so happened exactly one week ago. Houston Rodeo, a three-week major event, was cancelled after nine days on March 11 because of the coronavirus. Before cancellation, 8,000 petitioned online to cancel amidst Covid-19. Day after cancellation, 20,000 went online to petition Rodeo re-open. Perhaps it reflected a Texas state of mind, or perspective on the risk of being infected, i.e., “abundance of caution” vs “herd immunity”.

What is clear is that there is not a single surefire answer. Response really depends on the risk tolerance and where you land on risk-reward trade off. That is not a theoretical question amidst the oil price plunge to the lowest level in eighteen years.

Food for thought on machine learning applications:

1) tackling the Parent-Child well interference (frac hit), and 2) while we cannot predict commodity prices, how about closing the “gap” between expectation and actual production on Shale wells?

Call for action - as a community, lets do what we do best, e.g., share relevant experience in this TIG, and together better enable the industry to thrive in volatility and lower for longer WTI.