PSI - Issue 83

Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 83 (2026) 162–170

© 2026 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the ESIAM26 organizers Keywords: Predictive maintenance, crack detection, Euler–Bernoulli beam, Functionally graded material, Particle Swarm Optimization ; Abstract This paper conducts a comprehensive investigation into the utilization of the Particle Swarm Optimization (PSO) algorithm for crack detection in beam structures. The primary aim is to precisely forecast both the size and location of an open-edge crack in functionally graded beams. The chosen modeling approach involves a rotating spring, with its stiffness determined by the crack's size. The PSO algorithm is employed as an optimization technique to address this problem. The objective function is formulated as the weighted sum of squared errors between measured and calculated natural frequencies. The PSO algorithm systematically explores the solution space to identify optimal values for crack size and location, minimizing the defined objective function. The obtained results illustrate the efficacy of the PSO algorithm in accurately predicting crack size and location. This methodology provides a viable alternative for addressing crack detection challenges, opening avenues for potential applications in predictive maintenance of beam structures. In conclusion, the application of the PSO algorithm underscores its relevance and effectiveness in resolving crack detection issues, offering a promising solution for safeguarding structural integrity and averting future failures. The fourth European Conference on the Structural Integrity of Additively Manufactured Materials (ESIAM26) Advancing predictive maintenance: An optimization approach for detecting cracks in Functional Gradient Beam Structures Amal Lahrizi a *, Ayad Ghassane a , Abdelhamid Zaki b a Laboratoire Mécanique, Matériaux et Thermique (LMMT), École Nationale Supérieure des Mines de Rabat (ENSMR), Mohammed V University in Rabat, B.P. 753, Agdal, Rabat, Morocco b Laboratoire de recherche Artificial Intelligence & Complex Systems Engineering (AICSE), ENSAM-Casablanca, Université Hassan II, B.P. 20670, Casablanca, Morocco

* Corresponding author. Tel.: +212 7 66 66 02 68. E-mail address: a.lahrizi@enim.ac.ma

2452-3216 © 2026 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the ESIAM26 organizers 10.1016/j.prostr.2026.07.019