Application of neural networks and artificial intelligence tools for modelling, characterization, and forecasting in materials engineering

In the field of materials engineering, development of models for the characterization and forecasting of materials play a critical role in reducing the amount of experiment to be carried out by establishing a relationship between input and output. The traditional statistical prediction method can offer suitable output but often exhibit poor performance when the data given is extremely nonlinear. Nevertheless, the emergence of artificial intelligence and enhanced computing methods and tools have shown great capacity in addressing limitation. Therefore, this article presents several research works that have being carried out in developing highly accurate models utilizing soft computing methods like, genetic algorithm, random forest, finite element analysis, fuzzy logic, artificial neural network and gene expression programming for characterizing different categories of engineering materials (i.e. metals and their composites; polymers and their composites; ceramics and their composites).