Small plastic packaging films make up a quarter of all packaging waste generated annually in Austria. As many plastic packaging films are multilayered to give barrier properties and strength, this fraction is considered hardly recyclable and recovered thermally. Besides, they can not be separated from recyclable monolayer films using near-infrared spectroscopy in material recovery facilities. In this paper, an experimental sensor-based sorting setup is used to demonstrate the effect of adapting a near-infrared sorting rig to enable measurement in transflection. This adaptation effectively circumvents problems caused by low material thickness and improves the sorting success when separating monolayer and multilayer film materials. Additionally, machine learning approaches are discussed to separate monolayer and multilayer materials without requiring the near-infrared sorter to explicitly learn the material fingerprint of each possible combination of layered materials. Last, a fast Fourier transform is shown to reduce destructive interference overlaying the spectral information. Through this, it is possible to automatically find the Fourier component at which to place the filter to regain the most spectral information possible.

Separating copper from iron scrap is a critical operation in metal recycling and achieving this with low cost sensoric equipment like RGB cameras instead of XRF/XRT is becoming increasingly attractive. In this article, the groundwork for creating an image classification model to separate copper from iron scrap has been performed. Twenty of the most common and most easily available CNN architectures were trained on 2200 metal scrap specimens and evaluated inline on a sensor-based sorting rig for their prediction accuracy and their inference latency to mimic real circumstances in an industrial setting. Out of these evaluated architectures, DenseNet-201 with 98% accuracy in inline tests is recommended if potent hardware is available. Otherwise AlexNet with 92% accuracy or MobileNet-V2 with 90% accuracy are recommended for further investigation and model creation if hardware restrictions apply. Based on the presented results in this article, the initial cumbersome surveyance of the most suitable network architecture can be substantially reduced and the creation of a sorting model can be streamlined. This article thus provides the basis for creating an inline applicable sorting method for scrap metal that uses low cost sensorics equipment and can provide reasonably high accuracy in its prediction.

Improving the sortability of plastic packaging film waste (PPFW) is crucial for increasing the recycling rate in Austria as they account for 150,000 t of the annually produced 300,000 t of plastic packaging waste. Currently PPFW is thermally recovered, as it is impossible to separate the mechanically recyclable monomaterial films from the non mechanically-recyclable multimaterial films. In this study, machine learning models capable of classifying inline into monolayer and multilayer films of PPFW according to their spectral fingerprint taken in transflection were created. Feature selection methods, like PCA and MRMR F-Tests, identified the most relevant spectral ranges for classification, that show the least redundancy and highest relevance. This effective subset of features decreases the required complexity of the model while reducing prediction time without compromising accuracy. The resulting models achieved a prediction accuracy of 85 % on unseen specimens with minimal prediction latency, effectively showing the inline applicability of these models in sorting aggregates.

Meeting the recycled content targets of the Packaging and Packaging Waste Regulation requires significant improvements in recyclate quality for high‐performance rigid polypropylene (PP) packaging. However, the post‐consumer rigid PP stream is highly heterogeneous, resulting in intermediate recyclate properties when processed as a single stream. This study provides a comprehensive characterization of 20 representative rigid PP packaging products from an Austrian sorting facility, combining rheological, mechanical, thermal, molecular, and chemical analyses to derive a product‐property matrix and sorting implications. The results reveal substantial variability in melt flow rate, stiffness, impact strength, crystallinity, and molecular weight distribution, driven by polymer design and degradation during primary processing. Additional heterogeneity arises from chemical loads, particularly between food and non‐food products, as well as inconsistent stabilizer packages and barrier layers. These findings highlight inherent limitations of object‐based sorting, which cannot reliably infer material properties from external product attributes. Conversely, correlations between spectral‐relevant properties underscore the potential of property‐based flake sorting to classify materials by intrinsic characteristics. A systemic combination of both strategies is proposed: isolating food‐contact materials at the object stage and generating application‐specific recyclate fractions through flake sorting. Developing robust spectral‐property prediction models for PP remains a critical step towards substituting virgin grades.

The current transition from conventional blast furnaces to electronic arc furnaces is a viable path to reducing CO₂ emissions during steel production. However, this transition of technologies changes the requirements for possible scrap that may be used as a secondary raw material during EAF steel production. Copper is especially challenging, as it remains in the melt, reducing the mechanical properties of the produced crude steel while being lost to any secondary use. Currently, the two main routes to reduce the copper content are X-Ray Fluorescence sorting and manual sorting. We propose a third approach by using computer vision and machine learning methods to detect copper-containing particles in a post-shredder scrap fraction on low-cost and low-powered hardware. Furthermore, this proposed method is robust to environmental factors, such as heavily corroded particles caused by prolonged storage without proper weather protection. This method can effectively reduce the need for expensive XRF equipment or manual sorting. The developed sorting pipeline was examined in an industrial setting through sorting trials and achieved 99.9 wt.% purity in the produced iron fraction at throughputs of over 3 t/h.