INTEGRATING ARTIFICIAL NEURAL NETWORKS FOR ADVANCED SORTING AND SUSTAINABLE PROCESSING OF PLASTIC WASTE

Abstract

The improper disposal of plastic waste has indeed become a major environmental challenge contributing to pollution and sustainability woes. Conventional methods of sorting waste are not too effective, and as such, recycling is not so effective. In this, we propose an artificial neural network (ANN)-based system to upgrade plastic waste classification and preparation. Using the ANN model, we improve the recycling process by classifying the plastic waste by size and type and also by degradable and non-degradable materials. Plastic waste is transferred to a shredding machine with an automated conveyor system where rotating blades break it into smaller pieces. The mesh filtration process of the shredded plastic produces uniformly granulated granules. These are then redirected for further shredding into larger fragments. Thus, the refined plastic is further cleaned using water or chemical treatments to make it fit for reuse. The processed plastic is fed into the extruder, which molds the plastic into different types of products, e.g., tomato field sticks, dolls, and idols, depending upon the die used. Debris made of degradable waste is also used to produce biomass briquettes while organic debris is converted to fertilizers, securing a cycle economy along with eco-friendly waste management. The experimental results demonstrate that the ANN model increases sorting accuracy and processing efficiency much more than existing conventional methods. This research calls for a scalable and automated solution of sustainable plastic waste recycling through the use of AI-driven waste management techniques. The results demonstrate that the applicable ANN-based waste sorting systems can be successfully applied on a large industrial scale, enabling plastic waste treatment and improving the prospect of resource recovery.