Digital Twin is a technology that creates simulation models of machines or processes. When combined with sensing data, it enables real-time monitoring and analysis of the target system. However, challenges remain in creating high-precision models and in the associated costs. In this study, we propose a simplified Digital Twin based on 1D-CAE models generated through Model Based Development (MBD), for anomaly cause estimation. This approach does not produce a complete model of the target but rather a partial one. The model contains errors due to unmodeled factors and noise. Nevertheless, this method can estimate the causes of faults.

Accurate prediction of remaining useful life (RUL) is crucial for efficient maintenance of industrial equipment. Although various deep learning-based RUL prediction methods have been studied in recent years, effectively utilizing unlabeled data and long RUL data remains challenging. In this study, we apply a previously proposed method to the semi-supervised RUL prediction using time-series data. In this approach, a survival function modeled by neural networks is learned under hazard constraints. Introducing the survival function allows for consistent probabilistic handling of both labeled and unlabeled data. Experimental results on the CMAPSS dataset demonstrate that the proposed method outperforms baseline approaches.

In the development of human-robot collaborative systems, comprehensive risk assessment by designers and users is demanded. We previously proposed a risk assessment method that automatically identifies design information related to objective assessment by designers, using a dependency model of design information and a digital twin. In this study, the previous method is improved to take users' perception into account, using virtual reality. By developing a feasibility prototype of a collaborative conveyance system in a warehouse, the spatial design information related to worker's visual perception of a robot suddenly appearing from behind an obstacle was identified.

The concept of digital twin was proposed by M. Grieves in a PLM course in 2002. In 2019, Gartner named it one of the hot topics, and recently its applications have expanded beyond manufacturing to smart cities, healthcare, and more. However, due to this rapid expansion and development, its definition and interpretation have become unclear. This paper attempts to reorganize and clarify what digital twin is, its definition and conceptual framework with focus on the manufacturing industry, and reviews trends in standardization and implementation. It also introduces examples of research group activities and university education and human resource development.