Page 393 - Hydrocarbon Exploration and Production Second Edition

380 Risk Analysis for Major Capital Investments in Projects

BG
Concepts boundary conditions
+ inter-parameter relations
MS MD

n. alernative
stochastically-generated cases

x n.

n. equiprobable outcome
(with or without a base case)

probabilistic summary
Figure 15.17 Multiple stochastic approach to uncertainty handling.

created, each of which honours the fixed data specified. Each outcome is deemed
equiprobable, and the range of results (in say reserves) can be presented on a fre-
quency distribution, shown below, from which a p90, p50 and p10 case can be read.
The advantage of this technique is that it can generate a large number of possible
alternatives relatively efficiently, and will provide a fuller range than the base case-
dominated method. However, it is difficult to decompose the assumptions made in
any one case, for example the p90 outcome is just one combination of many input
uncertainties, and it is a less tangible combination than that offered by the third
approach, below.
The third approach is the multiple-deterministic method (Figure 15.18), in which a
limited number of discrete cases are built, by combining the key input variables
identified, in specific combinations. In the example below, the key variables are the
structural faulting and the reservoir properties. These are differentiated as discrete
concepts and then combined together in a limited number of combinations.
Alternatives can be narrowed down by dynamic modeling techniques such as
material balance, or well deliverability models which are ‘history matched’ against
observations made in the field. In the picture below the check is made against the