About
Akram Zarchini
Postgraduate Researcher
UCD
Akram Zarchini is a PhD researcher at I-Form, based at University College Dublin (UCD), working in the area of Smart and Sustainable Manufacturing. She obtained her undergraduate degree in Software Engineering from Sharif University of Technology. Her research focuses on integrating edge computing and machine learning to enhance sustainability in additive manufacturing (AM) processes. Her main interests include anomaly detection in AM, federated and graph-based learning, computational cost efficiency, and interpretable AI models for in-situ monitoring. She is particularly interested in applying intelligent, decentralized systems to improve the efficiency, sustainability, and reliability of AM technologies.
Technical Summary
The aim of Akram's work is to develop a smart and sustainable additive manufacturing (AM) system by enabling real-time in-situ process monitoring while minimising computational costs. Her research focuses on utilising edge computing and machine learning to monitor the printing process as it occurs, with the goal of detecting defects and irregularities such as energy fluctuations, porosity, or geometric inconsistencies during fabrication.
The study investigates how real-time data generated during printing can be processed locally—at the edge—to enable fast and reliable decision-making relying on cloud infrastructure. Special emphasis is placed on enhancing the transparency and interpretability of the monitoring system, allowing users to understand how process deviations are identified and to take informed corrective actions when necessary.
In-situ process monitoring is becoming increasingly essential in AM to ensure part quality and reduce waste. However, current solutions are often constrained by high computational requirements and latency. This research addresses these challenges by developing a lightweight, interpretable monitoring framework that delivers near-instant feedback on the manufacturing process. Ultimately, the goal is to develop more autonomous, energy-efficient, and cost-effective AM systems that contribute to the broader vision of sustainable and intelligent manufacturing.