This three-day training course focuses on the application of Artificial Neural Networks (ANN) to predict reservoir parameters, particularly in uncored wells. When combined with high-quality geological and petrophysical input data, ANN models can significantly enhance the predictability of reservoir properties and contribute to more reliable reservoir models. Strong geological and petrophysical control over data selection, handling, and interpretation is essential to achieve meaningful and robust results.

The course begins with a fundamental introduction to ANN concepts, followed by practical training using Tiberius ANN software and associated macros. Participants will learn how to gather and prepare data, run and validate models, interpret results, and identify common pitfalls in ANN applications. Through hands-on exercises, attendees will work through the complete workflow—from input data preparation to model development, testing, and interpretation. The models will then be blind-tested against a new well to assess predictive performance. Finally, the validated ANN models will be applied to uncored wells to enhance regional prediction of reservoir parameters.

By the end of the course, participants will have the knowledge and practical skills required to develop and run their own ANN models for reservoir characterization.