Circular Materials Informatics: Machine Learning for Polymer Sorting & Degradation

The aim of this workshop is to provide participants with a practical understanding of how machine learning can be applied to polymer sorting, degradation prediction, and circular materials management for sustainable recycling and waste reduction.

• To introduce the concept of circular materials informatics and its role in sustainable polymer management.
• To explain how machine learning can support polymer classification, sorting, and material identification.
• To demonstrate how data-driven models can be used to predict polymer degradation behavior.
• To help participants understand the importance of AI in improving recycling efficiency and circular economy practices.
• To provide exposure to practical workflows involving material data, feature analysis, and predictive modeling.

📅 Day 1: Circular Materials Informatics & Polymer Data Basics

Core Objective: Understand how polymer data and AI support sustainable recycling and circular material management.

• Introduction to circular materials informatics
• Polymer waste types: PET, PP, PE, PVC, PS, PLA
• Key polymer properties and recycling challenges
• Using MDPI research insights for polymer data understanding
• Basics of ML workflow for polymer informatics

🛠️ Hands-on Activity:

Polymer Data Explorer: Clean, visualize, and explore a sample polymer dataset in Google Colab.

🧰 Tools Covered: Google Colab, Python, Pandas, Matplotlib

📅 Day 2: Machine Learning for Polymer Sorting

Core Objective: Build ML models to classify polymers for smart recycling and automated sorting.

• Polymer sorting methods: spectroscopy, sensors, computer vision
• ML models for polymer classification
• Feature selection for sorting datasets
• Model evaluation using accuracy and confusion matrix
• Applications in recycling plants and smart waste systems

🛠️ Hands-on Activity:

AI Polymer Sorting Classifier: Train a simple ML model to classify PET, PP, PE, PVC, and PS.

🧰 Tools Covered: Google Colab, Python, Scikit-learn, Pandas

📅 Day 3: Polymer Degradation Prediction

Core Objective: Use ML to predict polymer degradation and support circular material decisions.

• Types of polymer degradation: thermal, UV, oxidative, biodegradation
• Key factors: temperature, humidity, UV exposure, additives
• Regression models for degradation prediction
• Predicting degradation percentage and material lifetime
• Applications in recycling, reuse, and sustainable material design

🛠️ Hands-on Activity:

Polymer Degradation Predictor: Build a regression model to predict polymer degradation percentage.

🧰 Tools Covered: Google Colab, Python, Scikit-learn, Matplotlib

• Understand the fundamentals of circular materials informatics.
• Explain the role of machine learning in polymer sorting and recycling.
• Identify key data types used in polymer classification and degradation studies.
• Understand how predictive models can support polymer degradation analysis.
• Recognize the real-world applications of AI in plastic waste management and sustainable materials development.
• Apply basic machine learning thinking to circular economy and materials science challenges.