The present work focuses on the study of metallic materials, specifically non-ferrous cast aluminum alloys. When working with these types of materials, it is considered important to have knowledge of their properties and characteristics, which largely depend on their chemical composition and the internal structure adopted by their atoms. An area of opportunity has been identified to improve the reliability and speed of obtaining results related to the quantification of physical and/or mechanical properties of metallic materials, particularly in processes that require experimental testing. Therefore, determining the values of these properties enables the control and prediction of material behavior when aiming to achieve or select specific characteristics. Based on the above, the importance of having access to a system capable of predicting the state of metallic materials used industrially in large volumes can be highlighted. Consequently, this work presents the use Artificial Neural Networks as a tool for predicting the values of hardness and elastic modulus from the chemical composition of cast aluminum alloys.