Internet of things (IoT) in digital circular economy principles, applications and challenges.
Atiokum, Alice (2024)
Atiokum, Alice
2024
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2024122037747
https://urn.fi/URN:NBN:fi:amk-2024122037747
Tiivistelmä
The Digital Circular Economy (DCE) emphasises sustainability by minimising various types of waste, including material waste, energy waste, and operational inefficiencies. It focuses on optimising resource use and extending the lifecycle of industrial assets, thus reducing the environmental and economic impacts associated with traditional linear models. The Internet of Things (IoT) has emerged as a transformative technology, enabling real-time monitoring and predictive analytics to achieve these goals. This thesis explores the implementation of IoT-based predictive maintenance within the DCE framework, focusing on leveraging sensor data and machine learning models for optimising industrial operations.
The literature review addresses two key research questions, examining the efficiency benefits of IoT in the “Digital Circular Economy” and the challenges IoT technologies aim to resolve. This highlights the potential of IoT to reduce resource waste and improve operational efficiency. It also addresses barriers like data interoperability, energy consumption, and scalability. The practical implementation involves analysing an open-access dataset from industrial air compressors. Using MATLAB, predictive maintenance models were developed, as guided by research question three. These models were designed to forecast equipment failures, enabling proactive interventions that align with DCE principles.
The results demonstrated high predictive accuracy across components, namely bearings, water pumps, and radiators, confirming the effectiveness of IoT-enabled systems in reducing downtime and improving the efficiency of resource management. Feature importance analysis revealed that operational parameters like motor power and noise levels play a key role in maintenance predictions, highlighting the effectiveness of IoT in improving predictive maintenance. However, challenges such as data imbalance, interoperability issues, and high energy demands were also identified. These challenges suggest the need for further advancements in IoT technologies to fully realise their potential in this field. Future research will focus on addressing these limitations to enhance system performance and scalability.
This experiment concludes that IoT-based predictive maintenance is a viable strategy for advancing DCE goals. The improvement of datasets, leveraging advanced machine learning techniques, and standardising IoT protocols will address the identified challenges and enable future implementations to enhance sustainability and operational efficiency. The findings highlight IoT's critical role in driving innovation within the circular economy, providing a pathway for sustainable industrial practices.
The literature review addresses two key research questions, examining the efficiency benefits of IoT in the “Digital Circular Economy” and the challenges IoT technologies aim to resolve. This highlights the potential of IoT to reduce resource waste and improve operational efficiency. It also addresses barriers like data interoperability, energy consumption, and scalability. The practical implementation involves analysing an open-access dataset from industrial air compressors. Using MATLAB, predictive maintenance models were developed, as guided by research question three. These models were designed to forecast equipment failures, enabling proactive interventions that align with DCE principles.
The results demonstrated high predictive accuracy across components, namely bearings, water pumps, and radiators, confirming the effectiveness of IoT-enabled systems in reducing downtime and improving the efficiency of resource management. Feature importance analysis revealed that operational parameters like motor power and noise levels play a key role in maintenance predictions, highlighting the effectiveness of IoT in improving predictive maintenance. However, challenges such as data imbalance, interoperability issues, and high energy demands were also identified. These challenges suggest the need for further advancements in IoT technologies to fully realise their potential in this field. Future research will focus on addressing these limitations to enhance system performance and scalability.
This experiment concludes that IoT-based predictive maintenance is a viable strategy for advancing DCE goals. The improvement of datasets, leveraging advanced machine learning techniques, and standardising IoT protocols will address the identified challenges and enable future implementations to enhance sustainability and operational efficiency. The findings highlight IoT's critical role in driving innovation within the circular economy, providing a pathway for sustainable industrial practices.