The increasing use of sensors and digital technologies in building management requires full integration of information into BIM models. However, few solutions exist that address this challenge organically using relational databases and open formats.This work develops the study of a framework for real-time monitoring of the Verona Arena (Italy) by integrating SHM data from sensors within an IFC model using SQL databases.The paper investigates how to relate different SQL databases of different building domains. Database integration provides a single environment for visualization and analysis of building conditions, improving maintenance and predictive management processes.
Retrofitting Europe’s ageing buildings is critical for sustainable development, yet traditional digitalization of architectural drawings, such as manual drafting, remains costly and inefficient. This paper presents a novel approach using Computer Vision (CV) to automate material detection and classification in drawings for simplified Material Passports (MP). Three convolutional neural network (CNN) architectures, including U-Net, ResNet50, and MobileNetV2, are evaluated, with MobileNetV2 outperforming for mixed-source data set despite some overfitting on well-represented materials. Case studies validate the model's practical effectiveness, especially for reinforced concrete elements, showing this CV-based method’s potential to aid digitalization and resource-efficient retrofits.
Material Passports (MPs) can play a crucial role in advancing Circular Economy (CE) principles, thanks to their features of documenting and tracking material properties throughout lifecycle. This study explores the potential of MPs as a valuable source of information for circularity assessments. A systematic review identifies key quantitative circularity indicators and their data requirements, which are compared with MP data availability. The resulting compatibility metric highlights MPs’ capability to provide essential material parameters for CE assessments. The findings prove the potential of MPs to support and simplify circularity assessments for enhanced effectiveness and greater accessibility.
Building materials heavily influence a building's environmental footprint. Effective material-related performance assessments demand knowledge of existing buildings’ characteristics, including material types, physical properties, and environmental indicators. However, structured data on material properties is scarce and often confined to closed repositories. While unimodal prediction models provide machine learning pipelines for predicting material properties, they overlook the intricacies of real-world scenarios requiring contextual insights from diverse data sources. This study introduces a multimodal building material property prediction method that incorporates building characteristics as contextual information. A residential building serves as the case study, focusing on brick facade to validate the proposed approach.
The conventional approaches for establishing indoor networks for pathfinding are mostly based on geometry processing. The generated networks are geometric networks that lack semantics, which have limited their use. This study proposes a new semantics-driven approach for constructing navigable indoor networks that utilizes the rich semantic information from Industry Foundation Classes (IFC) in the form of IFC-Graph. A real-world building model was used to validate the proposed approach, and the result shows that 1) the sematic-based indoor network is easier to establish from IFC-Graph; 2) the semantics-driven approach can enable semantics-driven pathfinding, which can provide more possibilities in pathfinding applications.
While Building Information Models (BIM) effectively store building-related information, accessing it requires specialized software and expertise. Natural Language (NL) interfaces for BIM data retrieval can mitigate this challenge, but existing approaches are limited by rigid ontological frameworks or extensive pre-processing requirements. We present a Large Language Model-based agentic workflow that processes NL queries and automatically interacts with IFC-encoded BIM models without ontological or pre-processing constraints. In tests across architectural, structural, and MEP domains, our approach achieves 80% overall accuracy. We provide open access to IFC-Bench-v1, our evaluation dataset containing various queries, answers, and reference BIM models.
This study investigates the impact of technology adoption on project management and field crew competencies in construction firms. A survey of 96 U.S. construction firms identified four clusters based on technology adoption patterns: Traditional Low-Tech, Selective Adopters, High-Tech Innovators, and Medium-Sized High-Tech firms. Firms with higher technology adoption performed better in project management areas like stakeholder relations, planning, and product acceptance, while field crew competencies remained consistent across clusters. These findings show the advantages of strategic digital integration and emphasize areas for addressing adoption barriers, providing practical guidance for firms navigating digital transformation.
In recent decades, awareness of climate change challenges has grown across engineering disciplines, as has the interest of researchers and practitioners in incorporating AI technologies into engineering solutions. Little has been discussed about the potential of AI in supporting societies to meet climate change challenges. This paper focuses on the understanding of AI regarding the need for climate resilient transport infrastructure. A framework setting a pathway to achieve this is proposed. Preliminary findings show that research in AI is being extensively developed for specific transport issues, but as a broader consequence, these can potentially increase the resilience of transport infrastructure systems.
Digital Twins (DT) has gained traction in recent years, offering significant potential to centralize design, construction and operation processes, improve management, and support decision-making. However, implementing DTs in the AECO sector involves managing various technologies and software, leading to increased complexity for professionals. This paper presents MatchFEM, a tool designed to simplify data generation for DTs within a unified environment. Developed as a plug-in for Grasshopper Computational Design (CD) software, MatchFEM facilitates 4D IFC-BIM modelling, IoT monitoring, data processing, and Structural Analysis models generation, key components of DTs. A case study is included to demonstrate its functionality and potential benefits.
Traditional methods for reviewing construction safety regulations are typically manual, time-consuming, and susceptible to inconsistencies. This study explores the use of Large Language Models (LLMs) to simplify complex regulatory language, thereby enhancing employers’ and workers’ understanding of OSHA policies and regulations. By processing images and textual reports from construction sites and regulations, LLMs can identify hazards, match them to relevant regulations, and provide actionable recommendations. This real-time, context-specific approach bridges the gap between regulations and practical application, fostering a safer, more informed workforce. Additionally, LLMs improve accessibility and comprehension of OSHA standards, aligning safety practices with regulatory requirements more effectively.
