Abstracts:Compositional reasoning, the cognitive process of breaking complex problems into manageable subproblems and recomposing them to generate new ideas, is fundamental to human problem solving and critical thinking. While deep learning models excel at pattern recognition, their capacity for true understanding and reasoning remains a topic of debate. Although the growth of the Internet has provided the necessary scale of data for model training, the way data is represented plays a pivotal role in enabling reasoning capabilities. This article introduces dynamic multimodal process knowledge graphs (DMPKGs), a novel neurosymbolic framework for data and knowledge representation that supports cognitive tasks such as compositional reasoning, high-level abstraction, explainability, and causal inference along with representation learning. The framework integrates data and knowledge into a unified, structured format enriched with semantics from multiple contexts. By prioritizing contextualized and semantically rich representations, DMPKGs aim to bridge the gap between pattern recognition and reasoning in artificial intelligence systems.

Abstracts:Digital twins (DTs) and artificial collective intelligence (ACI) are transformative technologies that, when combined, hold significant potential for managing complex systems across diverse domains, such as smart cities, health care, and manufacturing. DTs encompass both physical objects and their virtual counterparts, enabling real-time monitoring, control, and predictive modeling, while ACI enhances decision-making by leveraging the collective knowledge from multiple models. This article explores the synergies between DT and ACI, focusing on their integration into federated DTs (FDTs), which are networks of autonomous, collaborative DTs. By leveraging collaboration, FDTs optimize processes, improve scalability, and adapt to dynamic environments. We analyze the properties of DTs and ACI and identify opportunities for innovation and challenges in areas, such as scalability, adaptability, and fault tolerance. This integration paves the way for smarter systems capable of addressing the complexities of modern technological and societal challenges.

Abstracts:Digital twins (DTs) are becoming integral in sectors such as energy and manufacturing, catalyzing applications from monitoring and analysis to optimization and autonomous management. However, data in different formats, volumes, and qualities require high processing capabilities for integration. Moreover, the functionalities in DTs must work together, bringing interoperability challenges that have necessitated extensive manual programming. To address these, we propose the Internet of Twins, a disaggregated DT deployed in a distributed cloud architecture where interconnections are made using a remote procedure call framework, gRPC. Using this, we develop DT as a platform with knowledge-graph-based orchestration. Here, we implement a recursive execution algorithm to ensure necessary data are processed in the correct sequence autonomously. Then, we implement a wind energy pilot application and evaluate the performance. The results show that our approach lowers data preprocessing and line-of-code overhead up to 20% and 26%, creating a flexible and scalable architecture.

Abstracts:Digital twin (DT) technology is a promising solution for achieving optimized microgrid control with enhanced efficiency, reliability, and sustainability. In this article, we focus on a real-world microgrid in Singapore and develop a cognitive DT. Our DT consists of a client, located near the physical microgrid for real-time control, and a cloud-based server for running computationally intensive algorithms for energy management and optimization. We design and implement communication architectures to ensure seamless and real-time communication. The functionality and performance of our DT are validated through different microgrid-operational scenarios. The results show that our DT outperforms comparison algorithms significantly and approximates the theoretical optimal with merely a 0.24% difference in operation cost. Overall, we demonstrate the effectiveness of our DT in enabling real-time optimization and management of microgrid operations, paving the way for technology adoption in smart grids to achieve improved grid resilience and efficiency.

Abstracts:Although building information modeling (BIM) has been around for over five decades, its integration with Big Data generated by Internet of Things (IoT) networks has raised research interest. This fusion enables the digital twin of a building (DTB), a real-time virtual replica facilitating efficient location-based services, precise monitoring, and enhanced operational efficiency. Despite its alignment with the Smart City paradigm, widespread adoption is hindered by the initial investment for IoT network deployment. This article presents a comprehensive process, from optimizing IoT network deployment to implementing robust location-based services. A genetic algorithm leveraging the BIM database improves coverage by 44% compared to random deployments, while a multitier architecture based on deployed and roaming users’ devices extends service availability to 80% of devices without Internet access, enhancing overall service quality by ∼200%. Eventually, the versatility of DTB is showcased through the use case of indoor guidance services, concluding with potential research directions using the DTB.

Abstracts:The emergence of digital twins (DTs) has transformed domains like Industry 4.0 or automotive, enabling advanced insights and predictive maintenance as well as driving efficiency and innovation. With the evolution toward human-centric domains, such as Industry 5.0 or intelligent vehicles, the need for human DTs (HDTs)—human digital representations with the aim to align human interactions with the system’s design and its performance—arises. HDTs, linked to human interactions in various contexts, seek to improve well-being, safety, and productivity by integrating human factors into digital systems. However, they face unique considerations and challenges compared to traditional DTs. We provide an overview of distributed multicontext HDTs, highlighting their utility in various domains, and propose an architecture designed to address key challenges in their development and deployment, including data management, privacy, or interoperability. Additionally, we show the development of HDTs in practical and functional environments through a set of artifacts.

Abstracts:Plagued by both natural and human degradation, the historic Great Wall of China faces monumental challenges, limiting the effectiveness of current conservation methods. With its integrative and interactive analyzing capabilities, digital twin technology offers a promising solution. By enhancing planning and management strategies, improving risk monitoring and intervention, and providing multidimensional support, digital twin technology helps in the stability and possible reversal of the monument’s decay. Using the Beichakou Great Wall in Ningxia Hui Autonomous Region as a case area, this study proposes the “Beichakou Method.” Our research demonstrates the ways in which digital twin technology contributes to heritage conservation through heritage status assessment, conservation planning optimization, risk monitoring and intervention, presentation and transmission, and multifaceted system assurance. Despite these advancements, however, challenges remain in data acquisition, sharing, and coordinating multiparty interests, thus requiring further research.

Abstracts:Digital twin (DT) as a key enabling technology for achieving digitization, flexibility, and customization in shop floors has attracted significant attention. However, the shop floor involves diverse assets across multiple dimensions, scales, and interdisciplinary fields, making the modeling process complex. To address this issue, this article analyzes the construction process of ontology-based information models and proposes a generative modeling method for digital twin shop floor driven by large language models (LLMs). First, LLMs are utilized to analyze user intentions, acquiring the hierarchical object structure of DT models. Second, by combining an analysis–retrieval method to extract domain knowledge and generate dynamic prompts, LLMs are guided to realize the creation and fusion of objects and construct structured and semantically enriched DT models. Finally, the effectiveness of the proposed method is validated through examples of shop floor resource scheduling.

Abstracts:Digital twins (DTs) build the real-time digital mirrors of physical entities and play an important role in various industrial scenarios. The Internet of Things (IoT) serves as the backbone of collecting real-time data for building DTs to meet the technical requirements on real-time responsiveness and modeling precision. We propose a multifidelity data fusion (MDF) mechanism for digital twins via IoT, called MDF-DT. This mechanism establishes the digital twin of a physical entity by fusing real-time sensor data collected via IoT and historical finite-element method simulation data. An improved hierarchical regression for multifidelity data fusion (IHR-MDF) method is proposed to predict high-fidelity (HF) responses based on the low-fidelity samples taken from multiple sources and a small size of HF samples. Numerical experiments show that the normalized root-mean-square error is less than 0.4, and the computational time is about 0.2 ms/point. The proposed MDF-DT mechanism has high applicability in various DT applications.