The increasing need for multilayer packaging is the biggest contributor to the problem of plastic waste in the food and beverage sector. One of the types of multilayer packaging that is most often used as packaging is laminated foil. The low value of laminated foil in the recycling industry due to its high complexity makes a thorough evaluation of the use of laminated foil with the Life Cycle Assessment method necessary. The use of machine learning can be a user-friendly solution for calculating LCA scores. This series of research consists of a literature study, product data collection, dataset creation, and LCA standard formulation, and ends with algorithm development from a deep learning-based machine learning program. The literature study process produces important information which is then used for the process of collecting packaging data at retail places as well as making datasets involving Microsoft Excel and OpenLCA software. The building of machine learning programs then be done and becomes an important discussion in this research. To produce an optimal program, there are various parameters are tested. These parameters are test size, random state, number of layers and dense, learning rate, batch size, and epochs. This study resulted in a program with an accuracy of 97.09% which can be used to predict the LCA score of Laminated Foil packaging. © 2023 American Institute of Physics Inc.. All rights reserved.