Optical sensors and machine learning algorithms in sensor-based material flow characterization for mechanical...

Area of Science:

• Waste Management and Recycling
• Sensor Technology
• Artificial Intelligence

Background:

• Digital technologies offer untapped potential for enhancing waste sorting and processing plant performance.
• Improved sensor-based material flow characterization (SBMC) is crucial for advancing applications like adaptive plant control and sensor-based sorting (SBS).

Purpose of the Study:

• To provide a comprehensive overview of existing SBMC publications (2000-2021).
• To summarize current SBMC methods.
• To identify future research potentials in SBMC for waste recycling.

Main Methods:

• Systematic literature search of 198 peer-reviewed articles.
• Focus on optical sensors and machine learning algorithms for dry-mechanical recycling of non-hazardous waste.
• Analysis of publication trends and application evolution.

Main Results:

• SBMC research has significantly increased since 2019.
• Applications have expanded from solely sensor-based sorting (SBS) to material flow monitoring and process control.
• SBMC at material flow and process levels remains underexplored, with significant potential for upscaling from lab to plant.

Conclusions:

• Future research should integrate deep learning, low-cost sensors, and new sensor technologies.
• Utilizing data streams from existing SBS equipment can enhance SBMC.
• Advancements in SBMC can improve plant performance, increase material circularity, and support the circular economy.