Digital Twins in Smart Manufacturing and Healthcare: Bridging Engineering, IT, Law, and Management Disciplines

Abstract

Digital twin technology has developed into a revolutionary method in smart manufacturing and healthcare, facilitating virtual representation, real-time observation, and predictive modeling of physical systems. This study explores the creation and use of digital twins by combining engineering, IT, legal, and management viewpoints. Four algorithms—Random Forest (RF), Long Short-Term Memory (LSTM), Support Vector Machines (SVM), and K-Nearest Neighbors (KNN)—were utilized to model and forecast operational and patient results. Experimental outcomes show that LSTM attained the greatest predictive accuracy in both sectors, achieving 94% in smart manufacturing and 91% in healthcare, with mean absolute error (MAE) figures of 0.012 and 0.015, respectively. Random Forest demonstrated strong performance with 92% accuracy in manufacturing and 89% in healthcare, achieving a balance between accuracy and runtime efficiency. SVM and KNN, though useful with certain problems, also showed a little lower performance due to their susceptibility to high-dimensional or noisy data. Compared to previous studies, cross-domain tests and comparisons highlight the advantages of introducing multidisciplinary aspects, including regulatory compliance, data privacy, and the methods related to data management in enhancing the use of digital twins. The study identifies the potential of the digital twins to improve the industrial processes, advance the treatment of the patients, and support the strategic decisions, which preconditions the further investigation of the solutions in intelligent, resilient, and human-centred systems..