As the world continues fighting through this COVID-19 crisis, a question also emerges on which governmental responses are proving the most effective. Sweden notably did not shelter in place optimizing instead to develop herd immunity as quickly as possible. New Zealand took a different path hoping to eliminate cases by locking the country down, which has nearly eliminated new cases. Sweden on the other hand has not reached their end goal yet, and the health and economic numbers indicate they did not choose the wisest path.
Foreign governments make autonomous decisions and can consequently act as a testbed for solving common problems. The world observes the results and empirically judges the best policies and courses of action. This analog works exactly the same way in other contexts including the digital oil field (DOF). In the oil & gas (O&G) industry, there is a wide variety of prognostications and proffers on how best to capitalize on the promise of DOF including digital twinning and predictive maintenance. Industry participants willingly experiment by piloting and in some cases adopting new technologies that demonstrate sufficient value.
Production optimization has gone through the same experimental gauntlet. For decades, producers relied on manual data gathering and analysis tools to make production decisions. Although operations teams were not appropriately sized for large asset sizes, this approach was widely adopted as few options were available. Those early digitalization solutions pre-dated the deep operational cost-cutting measures and shrinking production teams we’ve seen over the last few years.
Prior to the downturn, a production engineer had responsibility for managing over 250 wells on average. That allowed for roughly 2 minutes each day to gather and analyze data, decide whether and what actions are warranted, and execute actions when appropriate on each well. Even the Michael Jordan of production engineering couldn’t operate at this level. In reality, engineers focus daily on only a subset of wells – typically 20-30. Is it, therefore, any surprise that 85% of all wells are in an under-optimized state today? The task of optimization became even harder with the downturn as cost-cutting measures and layoffs increased the number of wells an engineer must manage.
Never complacent, producers have experimented with different production optimization approaches over the last 5 to 10 years including better analytics tools, internal development or data science efforts, consulting-led custom development, and vendor-based artificial intelligence (AI). The experiment continues, but one approach has proven significantly more effective than others – especially in light of operational constraints. That solution is vendor-based artificial intelligence.
Over deployments across thousands of wells, Ambyint has seen up to 30% reduction in operating expenses and up to 7% increases in production volumes. Cloud-based computing and machine learning adeptly sift through vast amounts of data and execute important, repetitive tasks freeing engineers to focus on higher-value activities. AI is the very definition of scalability answering the question of how to optimize every well in a field with lean operations teams.
There is a perception that E&P lags other industries in innovation and technology adoption. While that knock has some merit given the appropriately conservative nature of engineers and the high capital requirements of many new technologies, E&P companies will adopt innovative solutions once they have shown consistent, needle-moving value. Fracturing technologies, for instance, saw rapid adoption once proven. With mainstream adoption, frac’ing increased productivity growth by more than 100% over a 5-year period.1 E&P is in fact only 1 of 4 industries to accomplish this feat since 1987.
Since advancements in drilling, innovation adoption has curtailed along with sector performance in the stock market. Even prior to the downturn, energy was the lowest-performing industry in the S&P. O&G embraces innovation when it changes the game. In tough economic times, E&P companies must resist the urge to retreat from innovative technologies and instead identify those advances that can achieve gains similar to ones in the past.
Artificial intelligence is on that shortlist of technologies. By assuming the repetitive tasks of analyzing data, detecting anomalous behavior, classifying wells, and driving control changes, AI gives E&P companies a fighting chance at not only optimizing their entire field of wells but maintaining that optimization state and positively impacting bottom-line performance. It also gives operations personnel back at least 25% of their day and reduces windshield time through a technology-enabled, management-by-exception model.
Countries are still evaluating the effectiveness of individual pandemic responses. With new data coming every day, the question still remains as to which countries are getting it right. What is clear is that as answers emerge, the rest of the world should quickly follow. There is no need to incur the excessive toll on blood and treasure that COVID-19 imposes when proven solutions are evident.
AI as a solution to scalable production optimization presents its own compelling evidence. E&P companies who follow others who have already embraced AI will avoid the unnecessary operating costs they incur today. In a market where cash flow is king, following this proven approach is strategically sound and for many a business imperative. If you have not adopted production optimization enabled by vendor-based AI, how much longer can you afford to wait?