What is the role of reservoir simulation in a big data world? This is not simply an abstract question. We think that the role of simulation in describing reservoir performance will change as big empirical data toolsets become democratized, expanding problem-solving approaches throughout an organization.
We are currently in a phase of technological flowering. New statistical methods can be conjoined to fundamental models of ever-greater power and predictive capability. Ironically, the power of new methods and ever-more-granular data often finds itself frustrated by usual common suspects: bad data, algorithms built without domain knowledge and failure to ask the right questions, among others. Regarding reservoir simulations and modeling, one important challenge is a simulation divorced from the empirical well result. Simulations that exist in outer space without a tie to reality have and will, sadly, break companies filled with smart people. Complex plays such as oil sands are examples, where inappropriate simulation reliance claimed Petrobank and Laricina.
Simulation alone is a beginning, not an end. It’s a question, not an answer. It’s a thought, not an action. The same can be said for empiricism. But when fundamentals are joined to statistics and domain knowledge, magic occurs. A model incorporating the multivariable impact of interwell spacing and proppant intensity on recovery exemplifies our contribution to the state-of-the-art:
RSEG continues to invest hundreds of millions of dollars building a platform that combines unique visualization tools, integrates big data and respects domain knowledge. In our view, these are minimum criteria to provide information that makes a difference to real time decision-making. Increasing organizational intelligence literally means making people smarter at every level. It’s easy to dismiss organizational conservatism as turf protection or not-invented-here syndrome, but the integration of big data throughout our own organization educated us on the paramount importance of domain knowledge and experience. Operating a complex enterprise requires thousands of micro-decisions based on evolved and often inarticulable rules. There’s a reason that field staff hate being dictated to by the head office.
And yet. We cannot fail to understand and deploy these new technologies as appropriate. Young and young-at-heart companies who embrace technology outperform incumbents with decades of experience. The first leg of this outperformance was based on a new application of industrial technology — big fracture stimulation. Subsequent legs will be based on intelligence technologies. We can see the difference between companies that engage in big data in order to check a box compared to those who engage in intelligence technologies that drive value. In the future, only one of these sorts of companies will exist.