Publication

Abstract : The wide use of synthetic packaging materials has led to serious environmental problems and the focus has been on the use of plant based packaging materials which are sustainable. Even though these biodegradable materials add greener alternatives, the most appropriate material in the packaging of individuals remains a complex question with the fluctuation of the material characteristics and requirements of products. In this research, we propose a recommendation system that can be driven by artificial intelligence and intelligently choose the most suitable plant-based packaging material based on the features of a particular product. The system uses domestic parameters, including biodegradability, mechanical toughness, flexibility and cost, to evaluate the compatibility of the material and product using machine learning and deep learning algorithms, like Feedforward Neural Networks, Siamese Networks and XGBoost Regressor. Results obtained through experiment prove that the Extra Trees Regressor produced the greatest R2 measure of 99.97%. In this work, we demonstrate how artificial intelligence can support the process of making data-driven, scalable, and sustainable decisions in the field of packaging design - a way of making industrial practices more sustainable.