Scholarlink

<p>Secondary raw materials are a&#xa0;key prerequisite for a&#xa0;functioning circular economy; however, their use is limited by technical, economic, and regulatory challenges. From a&#xa0;technical perspective, it is crucial that secondary raw materials exhibit sufficiently high and consistent quality in order to replace primary resources in demanding applications. Sensor-based sorting technologies play a&#xa0;central role in this context as they enable automated, rapid, and material-specific separation. This publication presents several case studies demonstrating how sensor-based sorting can be successfully applied to a&#xa0;wide range of tasks and waste streams.</p><p>(a)&#xa0;The sorting of post-consumer textiles is technically challenging. In the ‘StraTex’ project, AI-based methods were developed to evaluate feeding and singulation systems, along with datasets for second-hand textiles to enable automated, high-quality sorting.</p><p>(b)&#xa0;The ‘KIRAMET’ project improves the quality of steel scrap through AI-based detection and removal of copper-containing particles. The developed system achieves more than 99.9% separation efficiency and enables the use of the scrap in low-CO<sub>2</sub> steelmaking processes.</p><p>(c)&#xa0;In the ‘ReWaste&#xa0;F’ project, the Smart Waste Factory was implemented, enabling quality-driven process control through networked sensor systems and data platforms. For the first time, polypropylene from mixed waste was successfully recycled along the entire value chain into end products.</p><p>(d)&#xa0;‘ReSoURCE’ demonstrates the automated sorting of refractory waste using multi-sensor systems and AI, including small particle sizes. This results in purer fractions and new application possibilities as secondary raw materials.</p><p>In sensor-based sorting systems, the trend is clearly moving toward multi-sensor solutions and AI-supported, data-driven approaches. At the same time, there is still a&#xa0;need for standardized quality criteria, interoperable data and sensor interfaces, and economically viable solutions. Overall, the further development of sensor-based sorting represents a&#xa0;key lever for transforming waste management into a&#xa0;central, resource-efficient component of a&#xa0;functioning circular economy.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Sensor-based Sorting as a Key Enabler for Efficient Recycling and Recovery of Secondary Raw Materials—Insights from Case Studies

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月17日

查看全文
  • Alexia Tischberger-Aldrian,
  • Hannah Weber,
  • Gerald Koinig,
  • Alexander Egarter,
  • Renato Sarc,
  • Florian Feucht,
  • Philipp Sedlazeck,
  • Roland Pomberger

摘要

Secondary raw materials are a key prerequisite for a functioning circular economy; however, their use is limited by technical, economic, and regulatory challenges. From a technical perspective, it is crucial that secondary raw materials exhibit sufficiently high and consistent quality in order to replace primary resources in demanding applications. Sensor-based sorting technologies play a central role in this context as they enable automated, rapid, and material-specific separation. This publication presents several case studies demonstrating how sensor-based sorting can be successfully applied to a wide range of tasks and waste streams.

(a) The sorting of post-consumer textiles is technically challenging. In the ‘StraTex’ project, AI-based methods were developed to evaluate feeding and singulation systems, along with datasets for second-hand textiles to enable automated, high-quality sorting.

(b) The ‘KIRAMET’ project improves the quality of steel scrap through AI-based detection and removal of copper-containing particles. The developed system achieves more than 99.9% separation efficiency and enables the use of the scrap in low-CO2 steelmaking processes.

(c) In the ‘ReWaste F’ project, the Smart Waste Factory was implemented, enabling quality-driven process control through networked sensor systems and data platforms. For the first time, polypropylene from mixed waste was successfully recycled along the entire value chain into end products.

(d) ‘ReSoURCE’ demonstrates the automated sorting of refractory waste using multi-sensor systems and AI, including small particle sizes. This results in purer fractions and new application possibilities as secondary raw materials.

In sensor-based sorting systems, the trend is clearly moving toward multi-sensor solutions and AI-supported, data-driven approaches. At the same time, there is still a need for standardized quality criteria, interoperable data and sensor interfaces, and economically viable solutions. Overall, the further development of sensor-based sorting represents a key lever for transforming waste management into a central, resource-efficient component of a functioning circular economy.