Data-Driven Enterprise Operations: Hybrid Cloud Edge Data Architecture
Abdul Hannan, Hafiz (2025)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025120332251
https://urn.fi/URN:NBN:fi:amk-2025120332251
Tiivistelmä
The study addresses the challenges of integrating advanced technologies into industrial environments by developing an Industry 4.0 prototype. The prototype combines private 5G connectivity, edge computing with Nokia MXIE, Crosser data pipelines, the RXRM camera, and machine learning-based object detection using YOLOv8. The research aims to deliver harmonized, AI-ready data for real-time control while ensuring scalability, flexibility, and resilience in industrial operations. The theoretical framework incorporates IT/OT convergence, edge computing, and DataOps principles to optimize data quality, processing efficiency, and operational performance.
The methodology follows the Design Science Research approach. Implementation includes live-stream testing and video-based analysis to evaluate object detection and energy analytics work-flows. The system demonstrated robust performance in real-time monitoring, predictive mainte-nance, and operational optimization, highlighting its ability to address safety concerns and operational inefficiencies. The integration of machine learning and edge computing ensured low latency and actionable insights for industrial applications.
The results indicate the role of demonstration strategies in communicating the prototype’s capabili-ties to stakeholders. Process-focused demonstrations emphasized the integration of IT/OT components and the system’s potential to address aspects such as safety, scalability, and operational efficiency. The study suggests that the prototype aligns with industrial requirements and offers a solution adaptable to modern operational contexts within Industry 4.0 frameworks.
The methodology follows the Design Science Research approach. Implementation includes live-stream testing and video-based analysis to evaluate object detection and energy analytics work-flows. The system demonstrated robust performance in real-time monitoring, predictive mainte-nance, and operational optimization, highlighting its ability to address safety concerns and operational inefficiencies. The integration of machine learning and edge computing ensured low latency and actionable insights for industrial applications.
The results indicate the role of demonstration strategies in communicating the prototype’s capabili-ties to stakeholders. Process-focused demonstrations emphasized the integration of IT/OT components and the system’s potential to address aspects such as safety, scalability, and operational efficiency. The study suggests that the prototype aligns with industrial requirements and offers a solution adaptable to modern operational contexts within Industry 4.0 frameworks.