Search Results (1,885)

This study investigates the integration of Large Language Models (LLMs) into supply chain workflow automation, with a focus on their technical, operational, financial, and socio-technical implications. Building on Dynamic Capabilities Theory and Socio-Technical Systems Theory, the research explores how LLMs can enhance logistics operations, increase workflow efficiency, and support strategic agility within supply chain systems. Using two developed prototypes, the Q inventory management assistant and the nodeStream© workflow editor, the paper demonstrates the practical potential of GenAI-driven automation in streamlining complex supply chain activities. A detailed analysis of system architecture and data governance highlights critical implementation considerations, including model reliability, data preparation, and infrastructure integration. The financial feasibility of LLM-based solutions is assessed through cost analyses related to training, deployment, and maintenance. Furthermore, the study evaluates the human and organisational impacts of AI integration, identifying key challenges around workforce adaptation and responsible AI use. The paper culminates in a practical roadmap for deploying LLM technologies in logistics settings and offers strategic recommendations for future research and industry adoption. Full article

Automatic geodetic monitoring systems allow for real-time monitoring of an object’s condition. The article presents the IMSGeo system (Intelligent Monitoring System for Threatened Objects based on Automatic Non-invasive Measurements), which meets three fundamental efficiency criteria of a monitoring system: reliability, affordability, and the clarity of interpreted results. In this system, the surface is measured using reflectorless methods, and surface changes are determined based on the analysis of normal vectors. The studies were carried out for five typical surfaces: concrete, expanded polystyrene, tiles, brick, and metal. The experiment included two key aspects: analysis of measurement repeatability within accepted accuracy limits and analysis of geometry change determination using a proprietary algorithm. In the first case, a direct comparison of points was made using threshold alerts depending on the repeatability of the measurement. The differences generally did not exceed 5 mm. In the second case, the results showed that the maximum differences for brick and metal surfaces did not exceed 2 mm. For the polystyrene-covered surface, differences for 89% of measurements did not exceed 2 mm; for the tiled surface, 84% did not exceed 2 mm; and for the concrete surface, 97% did not exceed 5 mm. Full article

Buildings are among the most significant sources of energy consumption worldwide. Unfortunately, many are inefficient in terms of energy use, leading to high operational expenses. With modern technologies such as IoT sensors, smart meters, secure real-time communication, and advanced mathematical algorithms for data processing integrated into an efficient energy management platform, traditional buildings can be transformed into smart structures. In this context, a platform called “Building Energy Efficiency in Totality” (BENEFIT), which incorporates the smart building energy management (SBEM) concept, has been designed, developed, integrated, and tested as an innovative tool for monitoring and optimally controlling energy consumption. The platform is based on open-source software, enabling rapid and straightforward development of comprehensive solutions that address all aspects of the SBEM concept. The BENEFIT architecture allows the management of a wide range of devices within the building, including energy generation units, heating, ventilation, and air conditioning systems, indoor lighting, environmental sensors, surveillance cameras, and others. BENEFIT has been implemented and tested in a building belonging to the Faculty of Electrical Engineering at the Technical University of Iasi, Romania. The analysis of the results after one year of integrating the BENEFIT platform has resulted in a plan focused on measures to reduce energy consumption and improve the building’s performance and efficiency. The implementation of two measures (upgrading window insulation and improving lighting) resulted in a 12.14% reduction in total energy consumption. Full article

Masonry construction remains labor-intensive, with current block placement predominantly dependent on workers’ empirical knowledge. Lack of systematic cutting plans induces substantial material waste and rework, adversely affecting sustainability. We propose a two-phase optimization framework to automate and enhance masonry block arrangement efficiency. Phase 1 decomposes masonry structures into optimizable subregions by geometric features, documenting each region’s geometry and position to generate optimization datasets. Phase 2 implements a computational module using the Social Network Search (SNS) algorithm to optimize subregion layouts, recording post-optimization block coordinates and dimensions. Finally, it materializes layout configurations and generates block quantity schedules to provide precise material demand data. An integrated prototype system was implemented in four specialized block arrangement scenarios and one building case study, validating both functionality and efficiency. Full article

A systematic literature review is fundamental to building a robust research foundation, informing experimental methodology, and ensuring the quality of future scientific output. However, manual extraction of targeted information from scientific publications is often laborious and prone to error, especially when researchers require rapid access to relevant findings without specialized hardware. This paper introduces an automated workflow for information extraction from scientific publications in the engineering domain. The proposed workflow consists of two primary stages: data preparation and information extraction. During data preparation, PDF files are converted to plain text and segmented into logical sections using a rule-based block detection and classification algorithm for keeping semantics. Information extraction is then performed by applying regular expressions both on keys and values in the same sentence to identify and extract relevant process and material data from the segmented text. The approach was evaluated on a dataset of 18 open-access scientific publications from various journals and conference proceedings in the AM domain. The results of the automated extraction were compared with manual extraction and with a modern large language model (LLM)-based approach. The findings demonstrate that the proposed workflow can accurately and efficiently extract relevant process and material data, achieving competitive performance relative to the LLM-based method. The workflow offers a significant reduction in time and potential errors associated with manual extraction, with automated processing averaging 15 s per document compared to one hour for manual extraction, and achieving a 76% match rate. This efficiency enables researchers to rapidly and effectively extract data. The methodology is readily transferable to other scientific fields where systematic literature reviews and structured data extraction are required. Full article

In recent years, solar energy has emerged as a pillar of sustainable development. However, maintaining panel efficiency under extreme environmental conditions remains a persistent hurdle. This study introduces an automated defect detection pipeline that leverages deep learning and computer vision to identify five standard anomaly classes: Non-Defective, Dust, Defective, Physical Damage, and Snow on photovoltaic surfaces. To build a robust foundation, a heterogeneous dataset of 8973 images was sourced from public repositories and standardized into a uniform labeling scheme. This dataset was then expanded through an aggressive augmentation strategy, including flips, rotations, zooms, and noise injections. A YOLOv11-based model was trained and fine-tuned using both fixed and adaptive learning rate schedules, achieving a mAP@0.5 of 85% and accuracy, recall, and F1-score above 95% when evaluated across diverse lighting and dust scenarios. The optimized model is integrated into an interactive dashboard that processes live camera streams, issues real-time alerts upon defect detection, and supports proactive maintenance scheduling. Comparative evaluations highlight the superiority of this approach over manual inspections and earlier YOLO versions in both precision and inference speed, making it well suited for deployment on edge devices. Automating visual inspection not only reduces labor costs and operational downtime but also enhances the longevity of solar installations. By offering a scalable solution for continuous monitoring, this work contributes to improving the reliability and cost-effectiveness of large-scale solar energy systems. Full article

Artificial intelligence (AI) is now the computational core of smart building automation, acting across the entire cyber–physical stack. This review surveys peer-reviewed work on the integration of AI with indoor environmental quality (IEQ) and energy performance, distinguishing itself by presenting a holistic synthesis of the complete technological evolution from IoT sensors to generative AI. We uniquely frame this progression within a human-centric architecture that integrates digital twins of both the building (DT-B) and its occupants (DT-H), providing a forward-looking perspective on occupant comfort and energy management. We find that deep reinforcement learning (DRL) agents, often developed within physics-calibrated digital twins, reduce annual HVAC demand by 10–35% while maintaining an operative temperature within ±0.5 °C and CO₂ below 800 ppm. These comfort and IAQ targets are consistent with ASHRAE Standard 55 (thermal environmental conditions) and ASHRAE Standard 62.1 (ventilation for acceptable indoor air quality); keeping the operative temperature within ±0.5 °C of the setpoint and indoor CO₂ near or below ~800 ppm reflects commonly adopted control tolerances and per-person outdoor air supply objectives. Regarding energy impacts, simulation studies commonly report higher double-digit reductions, whereas real building deployments typically achieve single- to low-double-digit savings; we therefore report simulation and field results separately. Supervised learners, including gradient boosting and various neural networks, achieve 87–97% accuracy for short-term load, comfort, and fault forecasting. Furthermore, unsupervised models successfully mine large-scale telemetry for anomalies and occupancy patterns, enabling adaptive ventilation that can cut sick building complaints by 40%. Despite these gains, deployment is hindered by fragmented datasets, interoperability issues between legacy BAS and modern IoT devices, and the computer energy and privacy–security costs of large models. The key research priorities include (1) open, high-fidelity IEQ benchmarks; (2) energy-aware, on-device learning architectures; (3) privacy-preserving federated frameworks; (4) hybrid, physics-informed models to win operator trust. Addressing these challenges is pivotal for scaling AI from isolated pilots to trustworthy, human-centric building ecosystems. Full article

This paper explores how blockchain technology, widely known as the backbone of cryptocurrencies, can be harnessed to address limitations of traditional water quality monitoring (WQM) systems. Blockchain offers a decentralized, tamper-proof ledger that enables secure, transparent, and traceable data management across distributed networks. When applied to water quality monitoring, blockchain facilitates real-time data acquisition, enhances data integrity, and enables smart contracts for automated regulatory compliance and alerts. These features not only improve the accuracy and efficiency of WQM systems but also build public trust in the reported data. Key insights from current research and pilot applications highlight blockchain’s capacity to integrate with IoT devices for real-time sensing, support adaptive water governance, and empower local stakeholders through decentralized control and transparent access to information. The implications for policy and practice are significant: blockchain-based WQM can support stronger regulatory enforcement, encourage cross-sector collaboration, and provide a robust digital foundation for sustainable water management in smart cities and rural areas alike. As such, this review paper positions blockchain as a transformative tool in the digital transition toward more resilient and equitable water management systems. Full article

Conventional project scheduling techniques often rely on manual trial-and-error methods, which can lead to inaccurate evaluations. This study presents a dynamic scheduling framework to dynamically adjust scheduling decisions based on real-time productivity and budget constraints, resulting in improvement in scheduling accuracy in project management. By integrating advanced computational tools, the proposed approach addresses complex scheduling challenges. The model integrates Building Information Modeling (BIM)-based 3D data, productivity and process simulation, and optimization techniques to provide a unified scheduling tool that supports informed decision-making while considering real-time constraints, including productivity performance and budget limitations. The results demonstrated notable improvements over conventional methods, including a 13% increase in scheduling accuracy relative to the actual total project cost and a 34.4% improvement in scheduling accuracy based on the actual project duration, compared to the contractor’s baseline. The framework dynamically adjusts schedules and budgets according to current project conditions. These findings demonstrate its reliability as a decision-making tool for construction project management. The study introduces an integrative scheduling framework that adapts to real-time project conditions and is validated against actual project data. The integration of BIM, system dynamics, process simulation, and ACOR optimization provides a novel approach to construction scheduling. This methodology improves project management efficiency by automating scheduling adjustments based on ongoing progress. Full article

Optical wireless sensor networks (OWSNs) offer promising capabilities for high-speed, energy-efficient communication, particularly in mission-critical environments such as industrial automation, healthcare monitoring, and smart buildings. However, dynamic spectrum management and fault tolerance remain key challenges in ensuring reliable and timely data transmission. This paper proposes an adaptive spectrum management framework (ASMF) that addresses these challenges through a mathematically grounded and implementation-driven approach. The ASMF formulates the spectrum allocation problem as a constrained Markov decision process and leverages a dual-layer optimization strategy combining Lyapunov drift-plus-penalty for queue stability with deep reinforcement learning for adaptive long-term decision making. Additionally, ASMF integrates a hybrid fault-tolerant mechanism using LSTM-based link failure prediction and lightweight recovery logic, achieving up to 83% prediction accuracy. Experimental evaluations using real-world datasets from industrial, healthcare, and smart infrastructure scenarios demonstrate that ASMF reduces critical traffic latency by 37%, improves reliability by 42% under fault conditions, and enhances energy efficiency by 22.6% compared with state-of-the-art methods. The system also maintains a 99.94% packet delivery ratio for critical traffic and achieves 69.7% faster recovery after link failures. These results confirm the effectiveness of ASMF as a robust and scalable solution for adaptive spectrum management in dynamic, fault-prone OWSN environments. Full article

The increasing adoption of industrialized offsite construction (IOC) offers substantial benefits in efficiency, quality, and sustainability, yet presents persistent challenges related to data fragmentation, real-time monitoring, and coordination. This systematic review investigates the transformative role of artificial intelligence (AI)-enhanced digital twins (DTs) in addressing these challenges within IOC. Employing a hybrid re-view methodology—combining scientometric mapping and qualitative content analysis—52 relevant studies were analyzed to identify technological trends, implementation barriers, and emerging research themes. The findings reveal that AI-driven DTs enable dynamic scheduling, predictive maintenance, real-time quality control, and sustainable lifecycle management across all IOC phases. Seven thematic application clusters are identified, including logistics optimization, safety management, and data interoperability, supported by a layered architectural framework and key enabling technologies. This study contributes to the literature by providing an early synthesis that integrates technical, organizational, and strategic dimensions of AI-driven DT implementation in IOC context. It distinguishes DT applications in IOC from those in onsite construction and expands AI’s role beyond conventional data analytics toward agentive, autonomous decision-making. The proposed future research agenda offers strategic directions such as the development of DT maturity models, lifecycle-spanning integration strategies, scalable AI agent systems, and cost-effective DT solutions for small and medium enterprises. Full article

The accurate estimation of building heights using very-high-resolution (VHR) synthetic aperture radar (SAR) imagery is crucial for various urban applications. This paper introduces a deep learning (DL)-based methodology for automated building height estimation from single VHR COSMO-SkyMed images: an object-based regression approach based on bounding box detection followed by height estimation. This model was trained and evaluated on a unique multi-continental dataset comprising eight geographically diverse cities across Europe, North and South America, and Asia, employing a cross-validation strategy to explicitly assess out-of-distribution (OOD) generalization. The results demonstrate highly promising performance, particularly on European cities where the model achieves a Mean Absolute Error (MAE) of approximately one building story (2.20 m in Munich), significantly outperforming recent state-of-the-art methods in similar OOD scenarios. Despite the increased variability observed when generalizing to cities in other continents, particularly in Asia with its distinct urban typologies and the prevalence of high-rise structures, this study underscores the significant potential of DL for robust cross-city and cross-continental transfer learning in building height estimation from single VHR SAR data. Full article

In the IT sector, the relevance of looking at security from many different angles and the inclusion of different areas is already known and understood. This approach is much less pronounced in the area of cyber physical systems and not present at all in the area of building automation. Increasing interconnectivity, undefined responsibilities, connections between secured and unsecured areas, and a lack of understanding of security among decision-makers pose a particular threat. This systematic review demonstrates a paucity of literature addressing real-world scenarios, asymmetric/hybrid threats, or composite vulnerabilities. In particular, the attack surface is significantly increased by the deployment of smart sensors and actuators in unprotected areas. Furthermore, a range of additional hybrid threats are cited, with practical examples being provided that have hitherto gone unnoticed in the extant literature. It will be shown whether solutions are available in neighboring areas and whether these can be transferred to building automation to increase the security of the entire system. Consequently, subsequent studies can be developed to create more accurate behavioral models, enabling more rapid and effective analysis of potential attacks to building automation. Full article

The climatic conditions of a region are a constant object of study, especially now that climate change is clearly affecting quality of life and the way we live. The study of the climatic conditions of a region is conducted through meteorological data. Meteorological installations include a set of sensors to monitor the meteorological and climatic conditions of an area. Meteorological data parameters include measurements of temperature, humidity, precipitation, wind speed, and direction, as well as tools such as an oratometer and a pyranometer, etc. Specifically, the pyranometer is a high-cost instrument, which has the ability to measure the intensity of the sunshine on the surface of the earth, expressing the measurement in Watt/m². Pyranometers have many applications. They can be used to monitor solar energy in a given area, in automated systems such as photovoltaic system management, or in automatic building shading systems. In this research, both the implementation and the evaluation of an integrated low-cost pyranometer system is presented. The proposed pyranometer device consists of affordable modules, both microprocessor and sensor. In addition, a central server, as the information system, was created for data collection and visualization. The data from the measuring system is transmitted via a wireless network (Wi-Fi) over the Internet to an information system (central server), which includes a database for collecting and storing the measurements, and visualization software. The end user can retrieve the information through a web page. The results are encouraging, as they show a satisfactory degree of determination of the measurements of the proposed low-cost device in relation to the reference measurements. Finally, a correction function is presented, aiming at more reliable measurements. Full article

As large language models (LLMs) are increasingly integrated into software development, there is a growing need to assess how effectively they address subtle programming errors in real-world environments. Accordingly, this study investigates the effectiveness of LLMs in identifying syntax errors within large Python code repositories. Building on the bug in the code stack (BICS) benchmark, this research expands the evaluation to include additional models, such as DeepSeek and Grok, while assessing their ability to detect errors across varying code lengths and depths. Two prompting strategies—two-shot and role-based prompting—were employed to compare the performance of models including DeepSeek-Chat, DeepSeek-Reasoner, DeepSeek-Coder, and Grok-2-Latest with GPT-4o serving as the baseline. The findings indicate that the DeepSeek models generally outperformed GPT-4o in terms of accuracy (Acc). Notably, DeepSeek-Reasoner exhibited the highest overall performance, achieving an Acc of 86.6% and surpassing all other models, particularly when integrated prompting strategies were used. Nevertheless, all models demonstrated decreased Acc with increasing input length and consistently struggled with certain types of errors, such as missing quotations (MQo). This work provides insight into the current strengths and weaknesses of LLMs within real-world debugging environments, thereby informing ongoing efforts to improve automated software tools. Full article