Digital twin concept for manual waste sorting management

Global waste generation is projected to reach 3.40 billion tons by 2050, necessitating improved waste sorting for effective recycling and progress toward a circular economy. Achieving this transformation requires higher sorting intensity through intensified processes, increased efficiency, and enhanced yield. While manual sorting remains common, smaller plants often use positive sorting to recover recyclables, and larger plants combine automated systems with manual sorting. Negative sorting is employed to remove impurities and improve material quality. However, innovation in manual sorting has stagnated. Advances in Machine Learning and Artificial Intelligence offer transformative potential for waste management, with digitalisation and improved recyclate quality becoming priorities. Despite these trends, manual sorting is still largely treated as a digital black box. The presented research outlines the design of a novel, human-centric AI-powered assistance system to support sorting workers by enhancing decision-making and real-time assistance during the sorting process, driving the digitalisation of manual sorting. Potential use cases, system requirements, and essential components were explored. High-quality use case-specific data is essential for model training. Therefore, publicly available datasets were evaluated but found inadequate, necessitating use-case-specific data acquisition through near-industry-scale experiments. This data was used to train and develop key system components, such as object recognition, classification, and action recognition models. Results indicate that transfer learning with a balanced dataset is effective for waste-sorting applications. The classification model achieved 81% accuracy on an experimental acquired balanced dataset, outperforming the accuracy of the pre-trained model on its original dataset.

IntroductionIt is of upmost importance to contribute to fill the knowledge gap concerning the characterization of the occupational exposure to microbial agents in the waste sorting setting (automated and manual sorting).MethodsThis study intends to apply a comprehensive field sampling and laboratory protocol (culture based-methods and molecular tools), assess fungal azole resistance, as well as to elucidate on potential exposure related health effects (cytotoxicity analyses). Skin-biota samples (eSwabs) were performed on workers and controls to identify other exposure routes.ResultsIn personal filter samples the guidelines in one automated industry surpassed the guidelines for fungi. Seasonal influence on viable microbial contamination including fungi with reduced susceptibility to the tested azoles was observed, besides the observed reduced susceptibility of pathogens of critical priority (Mucorales and Fusarium sp.). Aspergillus sections with potential toxigenic effect and with clinical relevance were also detected in all the sampling methods.DiscussionThe results regarding skin-biota in both controls´ and workers´ hands claim attention for the possible exposure due to hand to face/mouth contact. This study allowed concluding that working in automated and manual waste sorting plants imply high exposure to microbial agents.

Innovations in manual waste sorting have stagnated for decades, despite the increasing global demand for efficient recycling solutions. The recAIcle system introduces an innovative AI-powered assistance system designed to modernise manual waste sorting processes. By integrating machine learning, continual learning, and projection-based augmentation, the system supports sorting workers by highlighting relevant waste objects on the conveyor belt in real time. The system learns from the decision-making patterns of experienced sorting workers, enabling it to adapt to operational realities and improve classification accuracy over time. Various hardware and software configurations were tested with and without active tracking and continual learning capabilities to ensure scalability and adaptability. The system was validated in initial trials, demonstrating its ability to detect and classify waste objects and providing augmented support for sorting workers with high precision under realistic recycling conditions. A survey complemented the trials and assessed industry interest in AI-based assistance systems. Survey results indicated that 82% of participating companies expressed interest in supporting their staff in manual sorting by using AI-based technologies. The recAIcle system represents a significant step toward digitising manual waste sorting, offering a scalable and sustainable solution for the recycling industry.

In this modern society due to rapid urbanization and developing technologies, the detection and segregation of plastic play a major role in human life. Currently, this task of segregating the plastic is manual and that requires higher effort in hazardous condition and it is dangerous too in some places (Gopinath et al. in International conference on technological advancements in power and energy, 2017) [1]. Due to the high rate of plastic consumption, it is important to develop a machine that can separate the plastic from normal waste material in regions wherein manual sorting is difficult to be followed and the segregation process should be efficient compared to the primitive techniques (Hussain et al. in 6th International colloquium on signal processing & its applications (CSPA), 2010) [2]. Currently, there are different types of methods implemented to segregate plastic such as manual sorting, post grinding waste sorting, optical waste sorting and floating waste sorting. In this project, a mechanical device is developed which contains a sensing unit that segregates the plastic based on audio wave signals that are produced during the crushing operation of the waste materials. The sensing unit will be pretrained to detect only plastic out of different element such as wood, steel, metal and plastic using machine learning technique. Mel frequency cepstral coefficients are used for plastic segregation, ultimately the proposed design of plastic segregation.KeywordsMachine learningMel frequency cepstral coefficientFeature extractionAudio feature extractionSensor unit servomotorActuation unit

Comparative analysis of three methods for estimating the compositions of construction waste.

Analysis of Picked up Fraction Changes on the Process of Manual Waste Sorting

Sorting and recycling of domestic waste. Review of occupational health problems and their possible causes

Robot-assisted automated sorting techniques for plastic recycling

Designing modern waste recycling systems is an urgent problem. The experience of attracting masters from two leading Russian universities in organizing interaction and cooperation in the development and implementation of waste sorting complexes is considered. Faculties of Russian Academy of National Economy and Public Administration and Moscow Institute of Physics and Technology held a project session for developing and launching innovative technological projects on the market called "Techno marketing". The developed methodological base made it possible to combine educational and entrepreneurial tasks and ensure the promotion of student technological projects. As a result of the conducted project session, a project on the use of computer vision systems for sorting waste was developed, the competitive differences of the project and its technical and economic indicators were presented. The mechanisms for the implementation of an innovative waste sorting project have been determined. Despite being twice as expensive, the waste sorting machine is able to fully pay off 3 times faster than manual sorting, justifying further investment in the project.

Within the framework of the European Green Deal, a circular economy is ensured, which is defined as a sustainable development model that preserves the value of products, materials and resources in the economy as much as possible. Waste sorting is of great importance in this process, as it enables rational management of resources and the return to circulation of already used products, which become raw materials for the production of other products. The proportion of unsorted waste in Latvia is higher than in other European countries, and the shared waste collection system operates with certain problems. In particular, this applies to the collection and management of bio-waste, which is currently one of the weakest sectors in the industry, despite the fact that the Landfill Directive of the European Parliament and the Council stipulates that a separate collection system for bio-waste must be in place by the end of 2023. In the event that the system of separate collection of biological waste is not organized during this period, sanctions may be applied to Latvia. Therefore, it is urgent to understand how to change consumer behaviour and increase involvement in waste sorting as the amount of waste increases, what opportunities and responsibilities exist for the organizations and consumers involved in this process. In this research, the study of consumer behaviour changes and their causes is analysed in the context of public communication and its opportunities. The aim of the article is to assess the importance of communication in encouraging changes in consumer behaviour in waste sorting in Latvia, especially in the bio-waste segment. In order to achieve the goal, an analysis of literature and documents was carried out on consumer behaviour, factors influencing it, consumer habits and communication possibilities for changing them. Consumer involvement and habits in biowaste sorting were investigated through a secondary analysis of previous relevant studies. In order to characterize the communication of waste management companies about waste sorting, a content analysis was carried out. In order to assess the current communication about waste management and sorting and its impact on consumer behaviour, interviews were conducted with communication experts. In general, it can be concluded that the waste management industry in Latvia is fragmented, which promotes competition between companies. However, each waste managing company has different waste sorting systems and other rules, which are not clearly explained to consumers, this communication is general and is not formed on a strategic basis, and in does not contribute to changes in consumer behaviour and an increase in waste sorting.

Occupational exposure during waste sorting is associated with several health outcomes. This study obtained knowledge about the impact of work in fully automated waste sorting plants (AWSP; n = 3) vs manual waste sorting plants (MWSP; n = 3) on personal exposure (n = 71) to bioaerosols and exposure-related health effects.Personal full-shift air samples were collected using various filter-based active sampling devices that were placed in the workers’ breathing zone. Personal exposure to inhalable and thoracic dust, endotoxin and microorganisms varied considerably between and within types of waste sorting plants (WSP). Workers at AWSP were on average exposed to 0.34 mg/m3 inhalable dust, 0.15 mg/m3 thoracic dust, and 51 EU/m3 endotoxins (geometric mean (GM) levels), whereas GM exposure levels at MWSP were 0.66 mg/m3 for inhalable dust, 0.44 mg/m3 for thoracic dust, and 32 EU/m3 for endotoxins. Exposure to submicronic fungal fragments did not differ between types of plants and ranged from levels below the detection limit (limit of detection, LOD) to levels in the order of 106 fragments/m3. Higher levels of fungal fragments and fungal spores were found at AWSP compared to MWSP with a GM of 2.1 × 105 spores/m3and with a GM of 1.2 × 105 spores/m3, respectively. Actinobacterial spores were found in samples from AWSP only, with exposure levels ranging from 1.9 × 104 to 1.1 × 107 spores/m3. Exposure to microbial DNA varied within and between WSP and was on average in the order of 104 copies/m3 for fungi and 105 copies/m3 for bacteria. Health symptoms, such as sneezing, congested nose and runny nose were significantly more common among exposed workers compared to the unexposed control group.

Urban mining of lithium-ion batteries in Australia: Current state and future trends

The orang utan. Its biology and conservation: Edited by Leobert E. M. de Boer. 1982. Dr W. Junk, The Hague. 376 pp. 24 × 15·5 cm. ISBN 90 6193 702 7. Price: Dfl. 175·00; US$76·00

Cyber-physical systems are defined by the integration of physical space entities and cyberspace information processing systems. Physical access control is generally perimeter-based, where assets can be vulnerable to a malicious entity once they have entered the perimeter of the space. Therefore, the relative distances between subjects and objects are needed to enforce cyber-physical access control within the perimeter of the physical space. The interplay between a physical entity and its virtual representation can be modelled using the concept of a digital twin. A digital twin enables the virtual monitoring of a physical entity to ensure better access control decision-making. This research presents a prototype indoor positioning and tracking system that can uniquely identify and track people and equipment in physical 3D space to create and maintain digital twins in real-time. The integration of 2D image processing and 3D depth-sensing technologies results in a system that can monitor a physical space where entities come into proximity to one another. Furthermore, the system can be used to prevent transgressions between physical entities within a relative distance of a few centimetres by tracking entities using human digital twin technology and reporting their relative proximity to an access control system for real-time enforcement.KeywordsCyber-physical systemsHuman digital twinsAccess control3D depth sensingStereovision

Toward a circular economy for plastics