As oil production demand increases, the industry as a whole faces the need to constantly renew its reserves by either discovering new fields or giving a second life to mature fields. The latter is particularly challenging for geoscientists and reservoir engineers. This paper presents a non-deterministic approach used to design a robust development plan for a complex mature carbonate field with more than a hundred wells and forty years of production history. This innovative approach can be applied to any redevelopment project for uncertainty assessment and risk management. The first step is a 2G&R synthesis performed by a multidisciplinary team to gather all the data, verify their consistency and assess the key heterogeneities and potential influential parameters to quantify the different production mechanisms. At this crucial stage, experimental design is extensively used. This synthesis enables a solid history-matched simulation case with an associated redevelopment strategy to be built efficiently. Secondly, the thorough knowledge of the field and the expected production mechanisms in the redevelopment phase are the key drivers to identify the most impacting static and dynamic parameters. These parameters are used as inputs for experimental design on both historical and redevelopment periods. Assisted History Matching techniques enable multiple simulation models honoring historical data to be found; the quality of these models is assessed by considering different objective functions. The history-matched models are then used to derive a reserves distribution for the project. Models outputs are combined to generate 3-D probability grids for properties such as oil saturation, in order to optimize the initial development plan and improve its robustness. This study highlights the importance of a strong and consistent 2G&R synthesis. An in-depth understanding of the field, combined with Assisted History Matching techniques, leads to models that honor the long production history, enabling the development scenario to be optimized.
