Applied Statistics

In the application of any mathematical method the need to know and understand the underlying assumptions cannot be understated. For an engineer, whether there is a need to satisfy the assumptions for a particular method will depend on the impact on the results.

Statistics is probably the most commonly used field of mathematics by technologists across many industry sectors. For sometime examples of their mis-use have been well documented.

There are established statistical practices associated with reservoir evaluation where due diligence to assumptions is warranted. For example

A further complication is that assigning probabilities is subjective;

....at the root of all uncertainty evaluations there is a need to assess probability in a subjective manner.

To soften this subjectivity views should be canvassed across a wide breath of experience such as in-house experts and joint venture partners, so that probabilities are assigned which have popular agreement

SPE Paper 26056

However, there is evidence in the literature to show it remains possible to omit  uncertainties or specify unrealistic probabilities despite collaboration in the specification of subjective probabilities.

The power of statistics in an industry where uncertainty can be managed but not eliminated is impressive. And the Geoscientists have used it to great affect in static modelling. For the reservoir engineer the paper;

Practical Geostatistics - An Armchair Overview for Petroleum Reservoir Engineers

SPE Paper 103357

provides some insight to the application of geostatistics.

However, for the engineer, Monte Carlo results generated on the basis of unrealistic assumptions could be potentially misleading . A view documented by David Nawrocki;

'The probability results from Monte Carlo simulation may look impressive to a client. However, if that number is derived from assumptions that are not realistic, there is no value to the number.'

The Problems of Monte Carlo