It is perhaps unfortunate that the word optimal is used so frequently in the industry with no indication of the sense in which it is to be understood. This can be particularly confusing in development programs when groups working on different parts of the program each consider their individual contributions to be in some way "optimal". The question that naturally arises is whether the overall plan is optimal.

From a restricted point of view the optimizing of a field development might be considered as a particular example of optimization theory, such as optimal control or mathematical programming. If, from the outset, an attempt is made to formulate it in such terms, as a well-defined problem for which solutions are readily available, it would both detract from our understanding of the problem and cast the solution in a misleading role.

Complex, real-world problems, such as field development, contain elements that are incapable of precise description and may even contain features that are contradictory. Such problems do not have the desirable properties and precise formulations that are necessary if the extensive literature on optimization is to be available for immediate application. In fact, for such problems, the literature is poorly provided with solutions. Therefore, it is necessary to examine the field development problem carefully to see how, and to what extent, quantitative solutions may be sought.

Clearly, we must address the problem of what is meant by "optimal" in the context of field development planning and show how it is related to the decision problems. Thus the relevance of objectives, criteria, constraints and information must be emphasised.

The type of decision dictates the information required and, therefore, to a great extent the type of model that may be used for evaluation.

Because field development planning Is essentially a dynamic problem, time is a prime factor. The prediction of behaviour, upon which decisions are based is a function of the state of knowledge of both the reservoir and the production mechanism and these in turn depend upon decisions made prior to the development decision and the time at which they were made.

In real development problems there will never be an adequate description of the system. Hence, all development decisions are decisions in the face of uncertainty. This may be recognized by the introduction of the concept of "fuzziness" which offers more realistic evaluation models.

You can access this article if you purchase or spend a download.