Delamination is one of the most common failure mode in laminated composites that leads the separation along the interfaces of the layers. The structural performance can be significantly affected by this degradation. Such damages are not always visible on the surface and could potentially lead to catastrophic structural failures. The existence of delamination alters the vibration characteristics of the laminated structures, so if they are detected and measured previously, they can be used as indicator for quantifying health and the potential risk of catastrophic failures. To ensure structural performance and integrity, accurate Structural Health Monitoring (SHM) is crucial. In this study, an optimized methodology for delamination identification on laminated composite plates involving the use of reduced mode shapes and computational tools, i.e., Genetic Algorithm (GA) and Artificial Neural Networks (ANN) is performed. In a first step, the sensor distribution on the surface of the structure was optimized using Fisher Information Matrix (FIM) criteria. After, GA and ANN were applied in order to identify and predict delamination location. A feed-forward based neural network is used in order to detect damage on the laminated plate using data obtained from Finite Element Analysis (FEA). The present methodology identifies damage localization in structures and also quantifies damage severity. The applicability of the technique is demonstrated on laminated plates and results are compared with numerical algorithms. This paper shows the effectiveness of GA and ANN as tools for delamination damage identification problem. The algorithms in their inverse formulations are capable of predicting accurately delamination position in plates-like structures.