Innovation: The automated plastic waste sorting technology utilizes Convolutional Neural Networks (CNNs) to recognize and categorize different types of plastic from images. A large dataset of labeled plastic images is used to train the CNN model. During training, the model learns to extract relevant features from the images using convolutional filters and pooling layers. The model's weights and biases are adjusted through backpropagation to minimize prediction errors. Once trained, the CNN model is integrated into waste collection businesses' mobile applications or waste management agencies' conveyor belt systems. When an image of a plastic item is captured, it is inputted into the model, which generates a prediction for the plastic's category. This prediction guides the sorting process, directing the plastic item to the appropriate recycling bin or container. By automating the sorting process, the technology eliminates the need for manual labor, reducing inefficiencies and errors. It significantly increases the accuracy and efficiency of waste recycling operations. The CNN model's ability to recognize and categorize the seven types of plastic enables real-time identification ,improving the overall efficiency and accuracy of plastic waste recycling processes and facilitating effective waste management and recycling practices.

Sector/Industry Application: Environment

Description: The waste recycling industry is a significant market with a global size of $47.2 billion. However, the sorting process remains a critical bottleneck for waste collection businesses and state-owned waste management agencies. This research project aims to develop and implement digital technologies that automate the plastic waste sorting process, benefiting waste collection businesses, such as Wecyclers and YoWaste, and government agencies like LAWMA. By leveraging artificial intelligence and integrating it into existing waste collection mobile applications, we aim to eliminate the need for manual sorting at sorting plants, resulting in a more efficient, cost-effective, and environmentally friendly waste management system. In summary, Cycle A.I's innovative solution addresses the critical issue of environmental pollution by automating the plastic waste sorting process. By eliminating the need for manual sorting, we streamline operations for waste collection businesses and state-owned agencies while significantly increasing recycling capacity. Our solution not only aligns with key SDGs but also contributes to creating sustainable cities, combating climate change, protecting marine life, and preserving land ecosystems.

Problem: The research aims at solving 2 major problems: 1. Inefficient and Manual Sorting Process: The traditional plastic waste sorting process is labor-intensive, time-consuming, and costly. Workers manually sort through the collected waste, leading to inefficiencies, errors, and high operational expenses. Cycle A.I's solution automates the sorting process using Artificial Intelligence, eliminating the need for manual labor and streamlining operations. 2. Limited Recycling Capacity: Waste collection businesses and government-owned waste management agencies face limitations in their recycling capacity due to the manual plastic waste sorting process. This restricts their ability to handle and process larger volumes of waste efficiently. By automating the sorting process, Cycle A.I's solution enables businesses and agencies to increase their recycling capacity, leading to more effective waste management and environmental conservation.