Jin-Woo Kwon

Physical AI enables reliable and timely operations of autonomous systems such as robots and smart manufacturing equipment under diverse and dynamic execution environments. In these environments, computing resources are often limited, shared among tasks, and fluctuate over time. This makes it difficult to guarantee that tasks meet timing constraints. As a result, resource-aware execution time prediction becomes essential for efficient resource management in physical AI systems. However, existing methods typically assume specific environments or static resource usage and often fail to generalize to new environments. In this paper, we propose CARE-D (Calibration-Assisted Resource-aware Execution time prediction), which trains a deep neural network to model the nonlinear relationships among hardware characteristics, resource levels, and task features across environments. The model predicts the execution time of tasks under diverse hardware and dynamically allocated computing resources, using a few execution records from new environments. CARE-D applies few-history-based calibration using only 1 to k execution records from target environments to adjust predictions without retraining the model. Experiments show that CARE-D improves prediction accuracy by about 7.3% over zero-history predictors within a 10% relative error and outperforms regression and deep learning baselines, using only one to five records per target environment.

Modern manufacturing demands real-time, scalable coordination that legacy manufacturing management systems cannot provide. Digital transformation encompasses the entire manufacturing infrastructure, which can be represented by digital twins for facilitating efficient monitoring, prediction, and optimization of factory operations. A Federated Digital Twin (FDT) emerges by combining heterogeneous digital twins, enabling real-time collaboration, data sharing, and collective decision-making. However, deploying FDTs introduces new concurrency control challenges, such as priority inversion and synchronization failures, which can potentially cause process delays, missed deadlines, and reduced customer satisfaction. Traditional concurrency control approaches in the computing domain, due to their reliance on static priority assignments and centralized control, are inadequate for managing dynamic, real-time conflicts effectively in real production lines. To address these challenges, this study proposes a novel concurrency control framework combining Deep Reinforcement Learning with the Priority Ceiling Protocol. Using SimPy-based discrete-event simulations, which accurately model the asynchronous nature of FDT interactions, the proposed approach adaptively optimizes resource allocation and effectively mitigates priority inversion. The results demonstrate that against the rule-based PCP controller, our hybrid DRLCC enhances completion time maximum of 24.27% to a minimum of 1.51%, urgent-job delay maximum of 6.65% and a minimum of 2.18%, while preserving lower-priority inversions.

As global energy demand intensifies across industrial, commercial, and residential domains, efficient and accurate energy management and control become crucial. Energy Digital Twins (EDTs), leveraging sensor measurement data and precise time-series forecasting models, offer promising monitoring, prediction, and optimization solutions for such services. Edge computing enables EDTs to deliver real-time management services placed closer to users. However, the existing energy management methodologies may fail to consider the limited resources of edge environments, which may cause service delays and reduced accuracy in management services. To solve this problem, we propose a novel energy control digital twin system with a resource-aware optimal forecasting mode selection scheme. The system dynamically selects optimal forecasting models by integrating statistical features of the input time series with available resources. It employs a two-stage approach: first, it identifies promising models through similarity detection in past time series; second, this initial recommendation is refined by considering the available computing resources to pinpoint the optimal forecasting model. This mechanism enhances adaptability and responsiveness in resource-constrained environments. Utilizing real-world LPG consumption data from 887 sensors, the proposed system achieves forecasting accuracy comparable to previous methods while reducing latency by up to 19 times in low-resource settings.

As user requirements become increasingly complex, the demand for product personalization is growing, but traditional hardware-centric production relies on fixed procedures that lack the flexibility to support diverse requirements. Although bespoke manufacturing has been introduced, it provides users with only a few standardized options, limiting its ability to meet a wide range of needs. To address this issue, a new manufacturing concept called the software-defined factory has emerged. It is an autonomous manufacturing system that provides reconfigurable manufacturing services to produce tailored products. Reinforcement learning has been suggested for flexible scheduling to satisfy user requirements. However, fixed rule-based methods struggle to accommodate conflicting needs. This study proposes a novel federated digital twin scheduling that combines large language models and deep reinforcement learning algorithms to meet diverse user requirements in the software-defined factory. The large language model-based literacy module analyzes requirements in natural language and assigns weights to digital twin attributes to achieve highly relevant KPIs, which are used to guide scheduling decisions. The deep reinforcement learning-based scheduling module optimizes scheduling by selecting the job and machine with the maximum reward. Different types of user requirements, such as reducing manufacturing costs and improving productivity, are input and evaluated by comparing the flow-shop scheduling with job-shop scheduling based on reinforcement learning. Experimental results indicate that in requirement case 1 (the manufacturing cost), the proposed method outperforms flow-shop scheduling by up to 14.9% and job-shop scheduling by 5.6%. For requirement case 2 (productivity), it exceeds the flow-shop method by up to 13.4% and the job-shop baseline by 7.2%. The results confirm that the literacy DRL scheduling proposed in this paper can handle the individual characteristics of requirements.

Cyber–physical systems have become critical across industries. They have driven investments in education services to develop well-trained engineers. Education services for cyber–physical systems require the hiring of expert tutors with multidisciplinary knowledge, as well as acquiring expensive facilities/equipment. In response to the challenges posed by the need for the equipment and facilities, a metaverse-based education service that incorporates digital twins has been explored as a solution. However, the issue of recruiting expert tutors who can enhance students’ achievements remains unresolved, making it difficult to effectively cultivate talent. This paper proposes a reference architecture for a learner-centric educational metaverse with an intelligent tutoring framework as its core feature to address these issues. We develop a novel explainable artificial intelligence scheme for multi-class object detection models to assess learners’ achievements within the intelligent tutoring framework. Additionally, a genetic algorithm-based improvement search method is applied to the framework to derive personalized feedback. The proposed metaverse architecture and framework are evaluated through a case study on drone education. The experimental results show that the explainable AI scheme demonstrates an approximately 30% improvement in the explanation accuracy compared to existing methods. The survey results indicate that over 70% of learners significantly improved their skills based on the provided feedback.

As natural disasters become extensive, due to various environmental problems, such as the global warming, it is difficult for the disaster management systems to rapidly provide disaster prediction services, due to complex natural phenomena. Digital twins can effectively provide the services using high-fidelity disaster models and real-time observational data with distributed computing schemes. However, the previous schemes take little account of the correlations between environmental data of disasters, such as landscapes and weather. This causes inaccurate computing load predictions resulting in unbalanced load partitioning, which increases the prediction service times of the disaster management agencies. In this paper, we propose a novel distributed computing framework to accelerate the prediction services through semantic analyses of correlations between the environmental data. The framework combines the data into disaster semantic data to represent the initial disaster states, such as the sizes of wildfire burn scars and fuel models. With the semantic data, the framework predicts computing loads using the convolutional neural network-based algorithm, partitions the simulation model into balanced sub-models, and allocates the sub-models into distributed computing nodes. As a result, the proposal shows up to 38.5% of the prediction time decreases, compared to the previous schemes.

Disaster management systems require accurate disaster monitoring and prediction services to reduce damages caused by natural disasters. Digital twins of natural environments can provide the services for the systems with physics-based and data-driven disaster models. However, the digital twins might generate erroneous disaster prediction due to the impracticability of defining high-fidelity physics-based models for complex natural disaster behavior and the dependency of data-driven models on the training dataset. This causes disaster management systems to inappropriately use disaster response resources, including medical personnel, rescue equipment and relief supplies, to ensure that it may increase the damages from the natural disasters. This study proposes a digital twin architecture to provide accurate disaster prediction services with a similarity-based hybrid modeling scheme. The hybrid modeling scheme creates a hybrid disaster model that compensates for the errors of physics-based prediction results with a data-driven error correction model to enhance the prediction accuracy. The similarity-based hybrid modeling scheme reduces errors from the data dependency of the hybrid model by constructing a training dataset using similarity assessments between the target disaster and the historical disasters. Evaluations in wildfire scenarios show that the digital twin decreases prediction errors by approximately 50% compared with those of the existing schemes.