Introduction:

The calculation of the possibility of success or risk of oil exploration in Cuba has not always been adjusted to its geological complexity, nor have appropriate methodologies been used. To solve this problem, the control parameters and geological processes for risk analysis were formalized and processed to reduce uncertainty, through a selection of statistical methods, fuzzy logic, knowledge engineering and geographic information systems, which allowed the adjustment and calculation of the potential of hydrocarbons in conditions of great geological complexity.

Methods:

The main existing methodologies and procedures applied worldwide were studied and their strengths and weaknesses were identified, which determined the strategy to be developed for the new application and its ability to manage knowledge.

Results:

The new application included for the first time, the probabilistic evaluation of the sedimentary basin and oil system, in addition to the plays, prospects and leads. More than 20 geological control parameters were created or modified, with the purpose of consistently estimating the probabilities of success in oil exploration and the reduction of uncertainty. The methodology was validated in the Guanabo Norte prospect, block E1B, located in the northern Cuban oil belt. The effectiveness of the methodology was demonstrated by comparing the results of the application (before drilling), with those obtained by the subsequent drilling of development wells. These determined a (deterministic) extractable reserve of 2,036 million cubic meters, while the methodology calculated the most probable extractable resources at 2,395 million cubic meters and a low probability of success for the prospect of 7.4%. This is considered a very good approximation. A web computing platform was created that facilitated and unified decision-making in oil exploration and where the foundations were laid to develop a knowledge analysis using machine learning.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )