Few analyses are more burdened with uncertainty than estimating the volumes of oil and/or gas that might be encountered in an exploration prospect, or group of prospects. It is not uncommon for the range of possible success-case volume outcomes to span several orders of magnitude.

In the past, explorationists often generated one map of a prospect and would make the assumption that the map constituted a 'best guess' as to what might be discovered. Post-drill 'lookback' studies have shown that such an approach was generally optimistic - often wildly so. Business decisions based upon this type of analysis often led to disastrous results.

In addition, the choice of which areas or geologic plays in which to explore and operate was often made based upon subjective or inconsistent information. This is an even more important business decision than prioritizing which prospects to drill. Most companies can tolerate the drilling of one or two dry holes, but operating in the wrong arenas can lead to financial ruin!

Most companies have come to realize the advantages of capturing the full spectrum of possible outcomes in analyzing undrilled prospects, or families of prospects. The problem is that many companies haven't had the resources to develop software to model the many different scenarios that they encounter in their day-to-day operations - for instance, multi-objective or amplitude-related prospects - and software packages available on the market are expensive.

Decision Strategies, Inc. has developed the industry's best suite of user-friendly, straightforward, and flexible software products to manage the risk analysis process in a proper and consistent way. Powered by Crystal Ball, the modules are designed to handle the assessment of chance and volumes for all common exploration scenarios.

Documentation in the form of a User's Manual, and embedded help/hint screens, remove many of the 'pitfalls' from the process and assure consistency of inputs. The screen below shows the embedded help from a portion of one of the modules.

Advantages of a Fully Stochastic Approach

Using a fully probabilistic methodology and Crystal Ball offers several advantages over software products that often rely upon a deterministic shortcut to a fully stochastic solution:

1. Partial dependencies, or correlations, between certain input variables (for instance, in some cases, productive area and pay thickness) can be easily modeled.
2. Input variable distribution types can be easily modified. For instance, while area and net pay distributions tend to be lognormal, the user has the ability to model triangular or uniform distributions for other variables, when warranted.
3. When analog data are available, Crystal Ball's Batch Fit tool can be used to determine which type of distribution best approximates the data.
4. In Play analysis, different thresholds can be modeled, for the minimum volume that must be found by your company early in your entry of the play to continue exploring - the 'play opening volume' - versus the minimum volume that has to be found in each discrete discovery to 'keep' that discovery -the 'prospect minimum commercial volume'. In addition, the user has the flexibility to model whether the 'play opening volume' must be housed in one discovery or can be contained by several discrete discoveries - the 'string of pearls'.
5. Mathematical shortcuts taken with deterministic methodologies can result in inconsistencies and errors when resultant distributions are truncated (for instance, due to a minimum commercial or economic threshold volume).

> Contact P. Jeffrey Brown of Decision Strategies, Inc. for more information.