Statistics and machine learning

Petroleum system analysis and geological modeling are complex tasks given the sparsity of data (well data, core interpretation…), the uncertainties on data interpretation (seismic interpretation, source-rock kinetic estimation…) and the time usually available during operational E&P projects. Although uncertainty management and risk assessment are key aspects of reliable portfolio management and project orientation, it is often basically performed using extreme modeling scenarios as most common ways to rigorously deal with uncertainties rely on massive multi-realization making unrealistic their use in operational projects.

C6+ consultants have developed solutions for uncertainty management and risk assessment

Thanks to their wide experience of petroleum system analysis C6+ consultants have designed solutions to accurately define key uncertainties and elements to be included into the risk analysis. their solutions give more systematic and robust results for efficient decision-making in E&P projects.. Therefore, they can provide powerful insight on the petroleum system and informative and quantitative risk assessments. The solutions they selected to perform these complex studies are based on machine learning technologies, making possible to perform the analysis in a time frame compatible with operational E&P projects, on computational intelligence, for providing geologically reliable results, and on statistical analysis for providing accurate outcomes.

Easily updating model predictions

Since a great part of the work is usually spend in building numerical models representative of the current knowledge on the studied petroleum system or stratigraphic model, C6+ developed powerful methods and workflows for easily and almost instantaneously updating model predictions when acquiring new data. Modeling results, and risks associated to the petroleum system, can be automatically updated when adding new data or interpretations without running any additional simulation. This is made possible by carefully taking into account uncertainties when constructing the model, making full use of the modeler’s expertise and experience.