The urgency of the discussed issue is caused by the need to develop the methodology for predicting efficiency of enhanced oil recovery methods. The methodology proposed can be used to evaluate the operation effectiveness as an alternative to application of hydrodynamic modeling, which does not always give accurate results at the high complexity of implementation in practice. The main aim of the study is to develop the methodology for predicting the technological efficiency of improving water injection profiles on injector wells by means of artificial neural networks and to check the predicting efficiency of the method developed on the basis of really conducted operations. The methods used in the study: analysis and summarizing of the results of improving water injection profiles on injector wells performed on one of the oil fields from 2008 to 2011; analysis of the influence of geological and physical characteristics and the technological productivity of wells on the total result of improving water injection profiles in terms of additional oil production due to lower water cut. The results. The paper demonstrates the possibility of using artificial neural networks for estimating the expected additional oil production as a result of improving water injection profiles on injector wells. Based on the operations of improving water injection profiles performed on one of the oil fields, the author has estimated the deviation in predicting the efficiency of improving water injection profiles by means of the suggested artificial neural network model. In comparison with the hydrodynamic modeling the developed mathematical model allowed obtaining forecast parameters for a shorter period with comparable prediction accuracy.