Artificial Intelligence (AI) is a broad discipline that uses powerful algorithms to emulate important aspects of human intelligence. Provided by the Industry 4.0 revolution, AI is increasingly applied in different fields from research to production. One of these fields is Materials Science and Engineering (MSE) which studies the relationships between processing, structure, properties, and performance of materials. The application of AI to MSE has triggered the invention of new materials to satisfy the demanding requirements in myriad sectors through the years. In this context, the “MSE paradigm” emerged as a framework to define these relationships supported by the available technologies at the corresponding time. This is how Polymer Matrix Composites (PMC) were synthesized. During the last years, they have turned from a futuristic solution to a necessity due to the wide range of advantages they offer concerning other conventional materials. The present work presents a modified approach to the MSE paradigm with the application of AI algorithms. An overview of the research advances from 2003 to 2019 in each fundamental link of the proposed MSE paradigm for PMC is exhibited in an organized fashion. This article must serve engineers and scientists working at the intersection of mechanical engineering, materials science and computer science to identify trendy topics in these fields. It aims to represent a starting point for developing innovative methods and proposing new research topics in the framework of the MSE paradigm powered by AI for PMC.

The present mathematical model for complex shells is given in the framework of Carrera unified formulation. The mechanical, electrical, and magnetic equations are derived in terms of the principle of virtual displacement, Maxwell’s equations and Gauss equations. Fourier’s heat conduction equation is used. The governing equations are discretized by the Chebyshev–Gauss–Lobatto and solved with the differential quadrature method. The three-dimensional (3D) equilibrium for mechanical, electrical, and magnetic equations are employed for recovering the transverse stresses, electrical displacement and magnetic induction. Finally, quasi-3D solutions for cycloidal shell of revolution and a funnel panel are introduced in this paper.

This paper presents Best Theory Diagrams (BTDs) employing combinations of Maclaurin, trigonometric, and exponential terms to build two-dimensional theories for metallic, multilayered, and functionally graded plates. The BTD is a curve in which the least number of unknown variables to meet a given accuracy requirement is read. The present refined models are Equivalent Single Layer and are implemented by using the Unified Formulation developed by Carrera. The plate theories presented are obtained using the axiomatic/asymptotic method and genetic algorithms. A multiobjective optimization technique is employed to analyze multiple displacements and stresses simultaneously. Closed-form, Navier-type solutions have been obtained in the case of simply supported plates loaded by a bisinusoidal transverse pressure. The influence of trigonometric and exponential terms in the BTDs has been studied for different materials and length-to-thickness ratios. The results show that the addition of such terms can lead to enhanced BTDs in which fewer unknown variables than pure Maclaurin expansions are needed to detect 3D like accuracies.