[EN] Concrete is the most widely used material in construction and the methods for its manufacture and dosage are quite popular in the world. In each of the regions of the world there are committees that define standardized requirements for the typification of concretes and one of the most relevant is the resistance to simple compression, this characteristic is used for requirements of large engineering works such as hydroelectric projects. The technology development in terms of materials, methods, and machinery has evidently advanced each time, such advances have gone hand in hand with the development of computer software and the use of advanced algorithms such as artificial neural networks (ARN), which have been applied in the satisfactory way of predicting results in areas such as biology, medicine, engineering, etc. In this master's thesis, a database obtained in the construction of a hydroelectric project in Ecuador will be used, and by using ARN, simple compressive strength will be obtained based on variables and parameters not normally taken into account for the design of dosage. Artificial intelligence, specifically ARN, has been used, since the results obtained are not of adequate quality through classical statistical methods. TFGM