An online knowledge leverage & sharing resource centre for engineering professionals and students in the field of reservoir engineering
Uncertainty Management
As Dake highlighted that uncertainty associated with physical mechanisms and processes, that a reservoir engineer hopes to understand and model, has been long recognised. As long ago as 1949 Muskat stated that;
"In its operational sense the principle of uncertainty, which is usually considered as limited to the realm of microscopoc physics, constitutes the very essence of applied reservoir engineering as a science"
Statistics have been used to capture that uncertainty within subsurface evauations for decades. Early applications saw the Monte Carlo Method used to generate probablistic reserves.
The SPE Paper 26056, a case study of the application of the 'Parametric Method' on the UKCS Harding Field, provides an illurstration the integration of uncertainty into reservoir performance predictons in 1993. At that time traditionally deterministic reservoir performance prediction were made based on a 'best technical' reservoir description.
The move towards probablistic rather than deterministic engineering has been driven by the recognition that the decision of field development approval should be made in full awareness of possible upside and downside outcomes. From this position downside mitigation strategies are possible.
The start of the 21st Century saw the start of rapid growth in the availability and use of reservoir simulators with probablistic and optimisation capabilities.The influence of uncertainty, if known, can now be evaluated with mathematical rigour.
However a note of caution. Statistics and probablistic theory are powerful methods when there is uncertainity but heed the words of Donald Rumsfeld (US Defence Secretary)
"....that is to say we know there are some things we do not know. But there are also unknown unknowns -- the ones we don't know we don't know"
If there are 'unknown unknowns' statistics and probablity methods are not going to change them into knowns. The unknowns unknowns have to be discovered, which is maybe what Dake was alluding to when he wrote;
The most valuable reservoir engineers are those who see the clearest and the most and who know what they are looking for.
The last comment implies the need for experience which may be a bit discouraging to newcomers to the subject but there has and always will be a very large element of déjà vu associated with reservoir engineering”
And it could be that those 'valuable reservoir engineers' are the ones who seek senergy and understanding across all the datasets strands that provide information concerning reservoir behaviour. As opposed to those who seek to minimise an 'objective function error'.
