SUMMARY
Comparing the quality of basic reservoir rock properties is a common practice to locate new infill or development wells for optimizing oil field development using reservoir simulation. The conventional technique employs a manual trial-and-error process to find new well locations, which proves to be time-consuming, especially for large fields. Concerning this practical matter, an alternative in the form of a robust technique is introduced in order to reduce time and effort in finding new well locations capable of producing the highest oil recovery. The objective of this research was to apply a genetic algorithm (GA) for determining well locations using reservoir simulation, in order to avoid the conventional manual trial-and-error method. This GA involved the basic rock properties, i.e. porosity, permeability, and oil saturation, of each grid block obtained from a reservoir simulation model, to which a newly generated fitness function was applied, formulated by translating common engineering practice in reservoir simulation into a mathematical equation and then into a computer program. The maximum fitness value indicates the best grid location for a new well. In order to validate the proposed GA method and evaluate the performance of the program, two fields with different production profile characteristics were used, fields X and Y. The proposed method proved to be a robust and accurate method to find the best new well locations for oil field development. The key to the success of the proposed GA method lies in the formulation of the objective functions.