Bhuiyan, Reyan Hridoy and Aesha, Umme Habiba and Khan, Sumaya Mahajabin and Jabeer, Muhammad Ahnaf (2023) AI-Driven Digital Twin Technologies for Advanced Manufacturing Systems. American Journal of Engineering , Mechanics and Architecture (2993-2637), 1 (10). ISSN 2993-2637

Text AJEMA_9646_Reyan+Hridoy+Bhuiyan_AI-Driven+Digital.pdf Download (1MB)

Abstract

With the fast evolution of Industry 4.0 the use of Artificial Intelligence (AI) and Digital Twin technologies have increased in order to improve the efficiency, reliability and intelligence of advanced manufacturing systems. A Digital Twin is a virtual model of a physical manufacturing resource used to monitor, predict and make decisions based on real-world operating data throughout its lifespan. The proposed research is an AI-based Digital Twin framework to realize advanced manufacturing systems, leveraging machine learning techniques and industrial sensor data to enhance equipment health monitoring and predictive maintenance. This study uses a dataset of 10,000 observations of milling machine processes collected by various sensors, including air temperature, process temperature, rotary speed, torque, tool wear, product type, and failure indicators of milling machines. Data is cleaned, transformed and prepared using a comprehensive data processing pipeline for building a model. Random Forest, Support Vector Machine, Decision Tree, XGBoost and Artificial Neural Network are all implemented and evaluated to find the best machine learning algorithm to predict machine failures. Conventional classification metrics including accuracy, precision, recall, F1 score, receiver operating characteristic – Area under the curve (ROC-AUC) and confusion matrix analysis are used to evaluate model performance. Overall, the proposed framework illustrates the potential for using AI-powered predictive analytics to complement Digital Twin technologies, by facilitating early fault detection, intelligent maintenance planning, and continuous optimization of operations. Digital Twin technology combined with AI helps minimize equipment downtime, enhance production reliability, schedule maintenance more effectively and boost manufacturing productivity. Also, the suggested framework enables data-driven decision making and offers a scalable structure for the deployment of intelligent manufacturing solutions in the context of Industry 4.0. The study builds on the ever-expanding body of research on AI-driven Digital Twins by providing a practical framework to boost manufacturing resilience, operational efficiency, and sustainable industrial performance, leveraging predictive intelligence.

Item Type:	Article
Uncontrolled Keywords:	Artificial Intelligence (AI), Digital Twin Technology, Advanced Manufacturing Systems, Predictive Maintenance, Machine Learning and Industry 4.0
Subjects:	H Social Sciences
Depositing User:	admin eprints
Date Deposited:	11 Aug 2026 23:39
Last Modified:	11 Aug 2026 23:39
URI:	http://eprints.umsida.ac.id/id/eprint/16918

Actions (login required)

View Item