We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Optimizing plastics recycling networks
AI summary Read the abstract
Researchers developed mathematical optimization models — including linear programming tools — to help plan efficient plastic recycling networks that can tolerate some contamination from mixed plastic waste streams. These models could help overcome a key barrier to large-scale recycling by intelligently matching waste sources with the plants best equipped to handle them.
Plastic pollution is a serious sustainability issue facing the global community. Fragments of macroplastics and microplastics pollute terrestrial and aquatic ecosystems, while nanoplastics can also degrade air quality. The recent COVID-19 pandemic also exacerbated the problem. Large-scale commercial use of plastics recycling technologies is hindered by various socio-economic barriers. In particular, cross-contamination of mixed plastic streams is prevalent due to imperfect waste segregation. The concept of Plastics Recycling Networks is introduced to facilitate planning of reverse supply chains using optimization models. In this work, basic Linear Programming and Mixed-Integer Linear Programming models are developed for matching sources of waste plastic with plastic recycling plants within Plastics Recycling Networks. These models allocate streams while considering the ability of recycling plants to tolerate contaminants. Two illustrative case studies are analyzed to demonstrate the effectiveness of the models, and policy implications for mitigation of plastic pollution are discussed. These models enable planning of networks with some tolerance for contaminants in plastic waste, and can be the basis for developing new variants to handle additional real world aspects. • Optimization models are developed for planning plastic recycling networks. • Linear programming and mixed-integer linear programming variants are formulated. • The models are demonstrated with two illustrative case studies. • Implications on effective management of plastic pollution are discussed.
More Papers Like This
Pinch-Based Synthesis of Plastics Recycling Networks
AI summary Read the abstract
This paper applied pinch analysis — an engineering optimization method originally used for energy systems — to design more efficient plastic recycling networks. The approach identifies where recycling streams can be combined or redirected to minimize waste and cost. The framework could help industries and policymakers design more effective plastic circular economy systems.
Process Optimization of Solvents Assisted Polyethylene Waste Recycling
AI summary Read the abstract
Researchers optimized solvent-based chemical recycling of polyethylene plastic waste to recover high-quality recycled material. The study demonstrates that chemical recycling can be tuned to maximize yield and quality, offering a scalable alternative to mechanical recycling that degrades plastic properties over time.
AI Techniques Aid for Optimizing the Collection System of Industrial Plastic Waste
AI summary Read the abstract
This study applied artificial intelligence techniques to optimize collection routes and predict demand for industrial plastic waste pickup. AI methods outperformed traditional statistical approaches in accuracy and route efficiency. Smarter collection systems could significantly reduce costs and improve recovery rates for industrial plastic waste.
Optimization of Renewable Energy Supply Chain for Sustainable Hydrogen Energy Production from Plastic Waste
AI summary Read the abstract
Researchers developed an optimization model for producing hydrogen energy from plastic waste in Iran using a renewable energy supply chain approach. They found that converting plastic waste into hydrogen fuel through gasification is economically feasible, with costs potentially decreasing over time as the system scales. The study presents a strategy for simultaneously reducing plastic waste volume and generating clean energy.
Efficient algorithmic coupling technique for precision recycling of seven types of mixed plastic waste
AI summary Read the abstract
Researchers developed a two-step machine learning coupling technique combining Linear Support Vector Classification (Linear-SVC) with a Multi-layer Perceptron (MLP) to improve the precision of sorting seven types of mixed plastic waste. The coupling technique raised overall plastic identification accuracy from 94.7% to 97.7% and substantially improved classification accuracy for HDPE and LDPE from 79-94%, while also reducing computation time compared to the single-step MLP approach.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.