Inadequate segregation of plastics and construction and demolition (C&D) waste drives carbon-intensive disposal and devalues resource streams, hindering high-quality recycling. Although artificial intelligence (AI) and machine learning technologies are increasingly applied to optimize waste management, most studies focus on operational efficiency or economic benefits, with limited integration of Life Cycle Assessment (LCA). This study addresses this knowledge gap by evaluating the research novelty of an AI-driven robotic storing system through a LCA. Results show that while robotic infrastructure contributes 18.1% to plastic waste impacts and 19.2% to C&D impacts, its efficiency yields significant net benefits. Compared to benchmark scenarios based on unsorted waste management, the system effectively offsets its manufacturing footprint by increasing recycling rates to 60% for plastics and 85% for C&D, thereby diverting waste from high-impact incineration, landfilling, and municipal solid waste management. Moreover, the assessments confirm that end-of-life (EoL) treatments remain the primary environmental drivers. In the plastic scenario, Climate Change (CC) is the most affected category (11.7 µPt), with incineration alone contributing 10.1 µPt. For C&D waste, the environmental profile is dominated by municipal waste treatment, particularly in human toxicity—non-cancer (HTnc) (2.79 µPt), where it accounts for 2.71 µPt. Additionally, CC impacts (2.25 µPt) are significantly influenced by both incineration and input waste transportation. Sensitivity analyses reveal that longer equipment lifespan, increased throughput, and higher recycling rates significantly reduce environmental burdens, whereas renewable energy use and shorter transport distances further enhance performance. Overall, integrating AI and robotics demonstrates strong potential to improve material recovery, reduce emissions, and support circular economy strategies for more sustainable waste management systems.
Parascanu et al. (2026) studied this question.