AI-Powered Anomaly Detection in Industrial IoT (IIoT) for Manufacturing Optimization
Fuente:
Zenodo
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Recurso digital |
| Publicado: |
Zenodo
2022
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866901778386124800 |
|---|---|
| author | Nirup Kumar Reddy Pothireddy |
| author_facet | Nirup Kumar Reddy Pothireddy |
| contents | <p><span lang="EN-US">The Industrial Internet of Things (IIoT) has changed manufacturing by allowing real-time monitoring of production lines and equipment efficiencies. However, sudden failures, factory slowdowns, and supply chain disruptions still undermine the operational stability. Traditional rule-based anomaly detection methods may create false positives and have no adaptability. In this study, an AI-powered anomaly detection system has been developed that integrates unsupervised learning (k-Means, DBSCAN) along with reinforcement learning to detect deviations in machine behaviors and processes.</span></p> <p><span lang="EN-US">The introduced model employs advanced analytics on IoT sensor data to then discover anomalies and optimize maintenance schedules through a recommendation engine. The authors experimented to prove significant reductions in 37% of downtime and 28% in maintenance costs, thus ensuring improved predictive maintenance and better operational efficiency. A case study further validates this research by discussing application to real-world problems. </span></p> <p><span lang="EN-US">Although it is effective, areas of improvement yet arise, like the issue of data quality and computational scalability. Further research will concentrate on improving the integration of deep learning and edge computing for real-time decision-making. The developed AI-driven self-adapting solution on the IIoT environment thus contributes to smart manufacturing.</span></p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_15202042 |
| institution | Zenodo |
| language | |
| publishDate | 2022 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | AI-Powered Anomaly Detection in Industrial IoT (IIoT) for Manufacturing Optimization Nirup Kumar Reddy Pothireddy anomaly detection machine learning unsupervised learning predictive maintenance manufacturing optimization <p><span lang="EN-US">The Industrial Internet of Things (IIoT) has changed manufacturing by allowing real-time monitoring of production lines and equipment efficiencies. However, sudden failures, factory slowdowns, and supply chain disruptions still undermine the operational stability. Traditional rule-based anomaly detection methods may create false positives and have no adaptability. In this study, an AI-powered anomaly detection system has been developed that integrates unsupervised learning (k-Means, DBSCAN) along with reinforcement learning to detect deviations in machine behaviors and processes.</span></p> <p><span lang="EN-US">The introduced model employs advanced analytics on IoT sensor data to then discover anomalies and optimize maintenance schedules through a recommendation engine. The authors experimented to prove significant reductions in 37% of downtime and 28% in maintenance costs, thus ensuring improved predictive maintenance and better operational efficiency. A case study further validates this research by discussing application to real-world problems. </span></p> <p><span lang="EN-US">Although it is effective, areas of improvement yet arise, like the issue of data quality and computational scalability. Further research will concentrate on improving the integration of deep learning and edge computing for real-time decision-making. The developed AI-driven self-adapting solution on the IIoT environment thus contributes to smart manufacturing.</span></p> |
| title | AI-Powered Anomaly Detection in Industrial IoT (IIoT) for Manufacturing Optimization |
| topic | anomaly detection machine learning unsupervised learning predictive maintenance manufacturing optimization |
| url | https://doi.org/10.5281/zenodo.15202042 |