Low Oil Prices and Big Data Analytics – What’s the Synergy?
The current challenging environment for oil prices has renewed the focus on improving efficiencies across all facets of E&P operations. Petroleum engineers and geoscientists are continually being asked to ensure that “faster, cheaper, better” strategies become standard operating practice. Unfortunately, a good understanding of the factors that are key to achieving these efficiencies is often lacking because of the interplay between complex geology and advanced engineering in today’s oil and gas development projects. This is where “Big Data Analytics” is increasingly being touted as a game changer. The narrative is that more and more sensors are being deployed to generate large volumes of data about the subsurface, the physical infrastructure and the flows. If we could get some additional insights about the reservoir by “mining” this data, then that could help increase the operational efficiencies.In this talk, I will discuss the premises, promises and perils of big data analytics by focusing on: (a) easy-to-understand descriptions of the commonly-used concepts and techniques, (b) broad categories of E&P problems that can be solved with big data analytics, and (c) case studies demonstrating practical applications. The first example to be discussed involves building robust predictive models for oil production in an unconventional reservoir using well architecture and completion data as predictors. The second example involves the ability to predict the presence or absence of vugular zones in carbonate reservoirs based only on a standard suite of electric logs. The third example involves building a data-driven model from historical injection-production data in waterflooding operations for optimization of injection rates and locations. The focus of the talk will be on showcasing an expanded repertoire of statistical and machine learning techniques that can help develop data-driven insights for understanding and optimizing the performance of petroleum reservoirs.