Delamination type of failure is extremely common in laminated structures and is a primary reason for failure in many use cases especially in aircrafts. In this project, Machine Learning Techniques were employed to detect the crack size and location in a four-layered laminated structure. The dataset for this problem statement was not available so it had to be generated and to do so ANSYS 18.0 was used. One structural model without any cracks was modeled and a training set with 2000 samples was generated to output the natural frequencies with different crack locations and sizes. Two regressor machine learning architectures with three algorithms (Linear regressor, Random Forest regressor and XGB Regressor) were developed for the prediction task, one to predict the area of the delamination and the other was a multioutput regressor model, which had to predict the X and Y coordinates of the center of the crack. The Random Forest Regressor gave the best generalizability in predicting the area of the delamination although linear regressor was not far behind as it performed quite remarkably given its simplicity. While predicting the locations, linear regressor gave the best test performance although hyperparameter tuning of the random forest and XGB regressor achieved similar results as compared with the linear regressor. © 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.