<p>The main objective of this paper is to provide an overview of the fundamental and practical aspects of machine learning and deep learning in durability analysis. First, it presents a wide range of applications of these technological developments in durability assessment and some of the challenges and limitations that hinder their broader adoption. In addition, a specific application of machine learning to predict the remaining useful life of lithium-ion batteries is discussed in detail, with a MATLAB implementation used to illustrate the computational nature of several deep neural networks. The results highlight that, despite their complexity, RNNs were outperformed by a simpler linear regression model. Finally, future research directions are proposed, and relevant conclusions are drawn from this review study.</p>