In this study, we would like to apply a Machine Learning (ML) algorithm that takes structural data and melt-pool lengths (MPL) of the structure to predict the new MPL, which we could use to simulate the model to increase the successful print rate. The machine learning algorithm we are using to predict the MPL is a regression model named Convolutional Neural Network (CNN) machine learning model. To achieve this, we are using data from the National Institute of Standards and Technology (NIST). With the provided MPL data from NIST [1], we created a 3D structure of the object based on the location where the laser has visited during the printing process. This structural data along with MPL of the points in the structure that are related to the target point are used to predict the MPL at that target point. We are also going to observe the effect of the structure that has already been printed on the MPL of the target point where the laser is pointed. This may give us the significance of the structure that has been printed in predicting the new MPL and anomaly occurrence. To achieve this, almost all the other parameters like speed of the laser, layer thickness, etc., are kept same throughout the printing of the object. The structure of an object that has been printed below the target point plays a major role in melt-pool depth occurred while the laser visits the point of interest.