Advancements in 3D digital model generation for digital twins in industrial environments: Knowledge gaps and future directions

Abstract

Digital twins are considered a transformative paradigm for industrial environments, providing a dynamic, digital, and intelligent representation of industrial assets. The necessity of digital twins in industrial settings is underscored by their ability to enhance asset monitoring, operational efficiency, and maintenance activities. The 3D digital model is fundamental for digital twins, serving not only as a digital representation of industrial environment but also facilitating the simulation of real-world scenarios. Although there have been extensive studies on the application of digital twins in industrial environments, the creation of 3D digital model for digital twins in existing industrial environments is still overlooked, primarily due to the complexity of these environments. This article aims to propose a workflow to create a 3D digital model for digital twins in existing industrial environments that includes four key components: 1) Data capturing, 2) 3D modeling, 3) Asset localization, and 4) Information integration. A significant body of literature on each component is surveyed to identify current knowledge gaps in harnessing 3D digital models for digital twins in industrial environments. In response to these gaps, this study proposes a series of future research directions, including automated data validation, real-time processing, semi-supervised or unsupervised learning-based 3D reconstruction methods, and 3D visualization approaches for industrial assets.