Video2Entities: A computer vision-based entity extraction framework for updating the architecture, engineering and construction industry knowledge graphs

Purpose The rapid advancement of unmanned aerial vehicles (UAVs) presents significant opportunities for the architecture, engineering and construction (AEC) industry. However, despite their potential, UAV applications face numerous barriers, including technical, regulatory and safety challenges, which hinder their widespread adoption. While existing research highlights these challenges, it provides limited guidance on comprehensive strategies to address them, which in turn may hinder the wider adoption of UAVs. This study seeks to fill this gap by identifying the multi-dimensional challenges associated with UAVs in AEC and developing a strategic roadmap that offers actionable insights for stakeholders, facilitating safer and more efficient UAV integration. Design/methodology/approach The 24 interviews were conducted to identify the challenges and potential strategies of UAV applications in the AEC industry. Thematic analysis was used to analyze the collected text contents, qualitative analysis was used to construct a conceptual strategy roadmap and focus group meetings were used to verify the roadmap. Findings The primary technical challenges include short battery life, loss of global positioning system (GPS)/compass and UAV magnetic interference. The critical safety risks are collisions, falls and absence of safety awareness among remote pilots and workers. Additionally, some practitioners struggled to meet operational requirements while maintaining the productivity of AEC flight operations. Furthermore, existing remote pilot training programs are generic and lack trade-specific training content. Research limitations/implications Despite the valuable benefits and contributions of this study, it is essential to acknowledge certain limitations encountered, which are outlined below. It is essential to note that the respondents in this study primarily came from Hong Kong, and the challenges may vary in different countries or regions. However, it is worth mentioning that the study interviewed specialists with international project experience who provided their insights from a global perspective. Practical implications This study contributes to a better understanding of the multi-dimensional challenges and strategies of UAV adoption in the AEC industry. The strategic roadmap will guide stakeholders to take proper actions for the safe integration of UAVs into the AEC industry. Originality/value This empirical study contributes to understanding stakeholders’ needs and engagement with regard to the new technology adoption. It also proposes a novel strategic roadmap for addressing the challenges through collaborative efforts among all relevant stakeholders, promoting safe and efficient UAV uses in the AEC sector.

Information and communication technology (ICT) has changed and continues to change almost every aspect of our daily lives. Our personal lives, social interactions, educational systems, business atmosphere, and corporate culture-everything has changed in one way or another due to advances in the field of ICT. The construction industry is no exception. ICT provides an unprecedented opportunity as well as enormous challenges to the construction industry.

Rule-enhanced iterative complementation for knowledge graph reasoning

The architecture, engineering and construction (AEC) industry is currently encountering numerous challenges and actively pursuing industrial upgrades through digital transformation. Digital twin (DT) technology aligns well with the industry's evolving needs due to its capabilities in data integration, intelligent decision-making, and effective management of project progress, efficiency, and quality. This study conducts a comprehensive literature review to analyze the strengths, weaknesses, opportunities, and threats (SWOT) associated with DT applications in the AEC sector. Additionally, field visits and semi-structured interviews were conducted on a selected construction project where DT was implemented, allowing for an in-depth examination of both its significant benefits and potential limitations. To further evaluate and prioritize the identified SWOT factors, a SWOT-AHP analysis was performed. The results indicate that the AEC industry has a strong interest in DT's advantages, particularly its potential to enhance resource allocation and process efficiency. Furthermore, the analysis reveals that the strategic quadrilateral occupies the largest area in the fourth quadrant (0.8554), with the strategic center of gravity located at (0.2321, -0.04135), suggesting that Strengths-Threats (ST) strategies should be implemented to support the future development and adoption of DT. Based on this strategic framework, the study formulates actionable development strategies to facilitate the effective integration and widespread adoption of DT in the AEC industry.

Aiming to solve learning difficulties caused by information explosion in the current era of education informatization, this paper proposes a precise teaching model driven by knowledge graphs. Knowledge graphs are essentially learning tools that effectively build a correct and complete curriculum knowledge system and accurately promote personalized learning paths. The main research contents are: (1) In view of the problem that the general knowledge graphs currently in use are not applicable in the field of education, we define the knowledge graph ontology structure of special courses based on Bloom’s teaching target system; (2) In view of the diversification of various teaching data sources, a simple knowledge graph representing structured data is used as a heuristic condition to extract a complete teaching sequence between knowledge points through self-expansion. (3) In view of the problem faced when trying to navigate relevant knowledge points in personalized learning, this paper exploits the internal relationship between knowledge point loopholes and ability achievements, and constructs an accurate path recommendation model based on quantitative data analysis. Taking the C language programming course as experimental data, this paper verifies the effectiveness of this model by quantitative means, which can significantly provide accurate teaching quality and realize the "multi-directional adaptation" among teachers, courses and students.

Large language models constitute a breakthrough state-of-the-art Artificial Intelligence technology which is rapidly evolving and promises to aid in medical diagnosis. In this study, we propose a novel evaluation framework for extraction of fine-tuning requirements based on the Objective Structured Clinical Examinations (OSCE) that can increase LLM potential and applicability. We developed an OSCE based evaluation meta-framework leveraging IoT-based data retrieval with a two-step approach designed to analyze and guide improvement of LLMs in multimodal medical diagnosis: (1) structured interaction evaluation and (2) domain-specific analysis of extracted data. Using Image-Metadata Analysis (IMA), Named Entity Recognition (NER), and Knowledge Graphs (KG), this framework identifies image domains, extracts relevant entities, and assesses connections in KGs. These methods collectively reveal areas for improvement, guiding fine-tuning to enhance diagnostic accuracy and contextual understanding in medical applications. Using this paradigm, (1) we evaluate the correctness and accuracy of generated medical diagnosis with publicly available multimodal-multiple-choice-questions in the vast domain of General Pathology and (2) proceed to the domain-specific analysis. We identify and visualize the model performance across specific organs, diseases, and pathological themes, detecting areas of lower accuracy, such as in cardiovascular conditions like atherosclerosis. This targeted approach enables precision-focused fine-tuning, applying additional data to specific weaknesses rather than a broad, generalized tuning across all pathology. Our framework’s primary contribution is its OSCE-inspired ability to dynamically identify and target under-performing areas within a broad domain, enhancing fine-tuning efficiency and diagnostic accuracy in a resource-effective and iterative manner removing the dependency on bulk adjustments, making it particularly suitable for sensitive applications where precision and resource efficiency are essential, such as in medical diagnostics. • Systematic Fine-Tuning Guidance: A structured two-step framework for identifying and addressing LLM weaknesses in multimodal medical diagnosis. • Multimodal Evaluation: Integrates textual and visual data to enhance real-world diagnostic accuracy, particularly in pathology. • Performance & Impact: Applied to GPT-4V, achieving 84% accuracy across 79 image-based medical multiple-choice questions. Targeted fine-tuning based on identified weaknesses is expected to yield substantial accuracy improvements. • Precision-Driven Enhancement: Uses Image Metadata Analysis (IMA), Named Entity Recognition (NER), and Knowledge Graphs (KG) to pinpoint areas for improvement, optimizing fine-tuning efficiency. • Broad Applicability: Adaptable across different LLMs and medical subdomains, with potential for generalization to other multimodal AI applications.

Rapid advances in artificial intelligence (AI) and machine learning (ML) are expected to bring major economic benefits. This benefit, however, also introduces the challenge of automating tasks central to many professions and redefining what consumers value. Despite extensive expertise and complex responsibilities, professionals in corporate real estate and the architecture, engineering and construction (AEC) industries are not immune to the effects of automation. A case study of the international AEC firm Ware Malcomb highlights opportunities from deploying AI and ML solutions for the corporate real estate and AEC industries by increasing efficiencies and value propositions. A broader trend to pursue short-term gains at the expense of long-term differentiation with AI and ML may, however, pose risks. Primarily, automation may reinforce commodification that has perniciously diminished the value of corporate real estate and AEC services. While achieving greater efficiency in project delivery is essential in the short-to-midterm, it is crucial to use the time and resources freed by automation to innovate new value propositions and real estate and design performance metrics that drive long-term innovation and growth.

PurposeThe unexpected spread of COVID-19 rapidly switched from a health crisis to an economic one. The Architectural, Engineering and Construction (AEC) industry experienced drastic impacts, especially in Africa. Several studies investigated COVID-19 impacts on the AEC industry, but very few were conducted in Africa. This study aims to cover this gap, address detailed overview of negative and positive impacts of COVID-19 on the AEC field, especially in the different African regions, and highlight their causes and the measures taken to overcome them.Design/methodology/approachThe authors combined a Preferred Reporting Items for Systematic Reviews and Meta-Analyses-based Systematic Literature Review (SLR) and a survey involving 87 AEC companies operating in Africa. The SLR initially used four scientific databases; however, considering the limited Africa-related found data, institutional and governmental databases were also included.FindingsGlobally, implementing the mandated restrictive measures against COVID-19 caused significant losses for developers, designers and contractors but helped the information and communication technologies operators to thrive. In the five African regions, the AEC industry experienced 22 heavy impacts that can be split into four categories: financial, managerial/strategic, operational and opportunities. This paper thoughtfully explains the causes of COVID-19 impacts and presents the undertaken measures by the African private and public sectors to overcome them. Generally, the African AEC industry lost 51% of the total sales in 2020.Originality/valueThis paper contains all aspects related to health hazard influences on the AEC industry, especially in Africa. Researchers and decision-makers may use it to build new approaches or strategies related to risk management or design technological solutions.

Cobb's Paradox (Bourne, 2011) asks: 'We know why projects fail; we know how to prevent their failure-so why do they still fail?' This study immerses itself into a major Australian IT project in order to unearth the drivers of project failure. Several new and novel findings have emerged. Using Multi-Grounded Theory this research has developed models and rich descriptions of new phenomena. The phenomena identified in this research, are drawn from social psychology and economic theory and highlight the issues of project execution as a social undertaking. This paper addresses one of those findings, namely the lack of domain expertise by senior management and vendor representatives. This paper examines the consequences of 'actors-working-in-organisations' (Manning, 2008, p. 678) and in particular looking at individual interactions, decisions and consequences (Goffman, 1959) through the lens of the Kruger-Dunning Effect (1999).

The building industry is a major consumer of natural resources and a large contributor to environmental degradation, leading to a need to rethink current building practices. Digital fabrication (Dfab) technologies, which transform design and engineering data into physical products, are gaining traction in the Architecture, Engineering and Construction (AEC) industry. This study aimed to evaluate the implications of digital fabrication in the construction industry, by identifying the current Dfab applications and the hindrances that are limiting its implementation. The research questions addressed were why Dfab is essential in the construction sector, the current state-of-the-art of Dfab in the construction industry, and how Dfab is improving the construction industry. Through a systematic literature review, the findings proposed that Dfab can revolutionize the construction sector, enabling freeform architecture, reducing construction costs, cutting material waste, and increasing worker safety. Nevertheless, further research is needed to overcome obstacles such as high costs and the lack of digital skills in the construction industry

Application of drones in the architecture, engineering, and construction (AEC) industry

PurposeThis study aims to investigate the interrelationships between COVID-19 pandemic impact and architectural, engineering and construction (AEC) organizations. To achieve that aim, the objectives are to: identify the critical pandemic impact (CPI) of COVID-19 on AEC organizations; compare the CPI between the different organization characteristics; and analyze the interrelationships between the CPI.Design/methodology/approachA systematic literature review of 519 articles and in-depth interviews with 40 AEC practitioners were carried out, yielding 12 COVID-19 impacts. Then, data on the 12 impact’s criticality were collected through a survey among AEC practitioners. In total, 127 valid responses were received and analyzed using mean ranking analysis, analysis of variance and Spearman’s correlation. The findings were also compared with pandemic impacts from prior works.FindingsFive CPI exist among the 12 impacts, namely, “reduced construction productivity,” “reduced foreign investment in the construction industry,” “reduced demand on construction-related works,” “disruption in the supply chain” and “reduced number of public projects.” “Reduced foreign investment in the construction industry” and “reduced demand on construction-related works” significantly impact small-medium enterprises than large enterprises. Also, “disruption in the supply chain” and “reduced construction productivity” are recurring impacts in all countries compared. Finally, moderate relationships exist between “reduced foreign investment in the construction industry” and “reduced demand on construction-related works” and “disruption in the supply chain” and “reduced construction productivity.”Originality/valueThe findings can be used to prioritize available resources in developing short-term responses for COVID-19. Also, action plans can be developed to mitigate the CPI from recurring.

The construction industry of Malaysia is very dynamic in nature and the adoption of new technological advancements and construction methods has been a prime concerns for its key players and stakeholders. However, the adoption of BIM in such a dynamic industry is observed to be limited and its implementation has not been as successful as it should have been. The study focuses on the identification and prioritization the factors challenging the implementation of BIM in Malaysian construction industry. The data has been administrated through designed questionnaire by identifying the factors / barriers, by literature review, for implementation of BIM in construction industry. The prioritization of such factors is expected to help the major stake holders to address the issues according to their priority which will save lot of previous time and financial with smooth implementation of BIM in Architecture, Engineering and Construction (AEC) industry in future project.

The significant environmental impact of the construction industry, its critical role in the economy, and the urgent need to respond to global climate change has created economic and environmental challenges that the architecture, engineering and construction (AEC) industry is increasingly encountering. The diversity of priorities and differing perceptions of risk about the adoption of green building practices among industry stakeholders is critically shaping the adoption of green technology, and the rate at which the industry is shifting towards more sustainable practices. In this paper, we develop an integrated framework that uses a probabilistic definition of value to integrate reductionist methods of sustainable decision-making, within a complex adaptive systems definition of the context within which such decisions are made. It is founded in probabilistic risk assessment and system dynamics, and is used to aid individual adoption of sustainable technology and study its diffusion through professional networks in the AEC industry.

Purpose The architecture, engineering and construction (AEC) industry is currently undergoing a paradigm shift as it integrates innovations such as digital twins (DT) in its activities. As a result, this study aims to ascertain the barriers affecting the implementation of digital twin (DT) technology in Nigeria’s AEC sector. Design/methodology/approach The study employed a quantitative approach using a questionnaire distributed via Google Forms, yielding 120 valid responses from built environment professionals in Nigeria. The data were subjected to statistical tests such as the Kolmogorov–Smirnov test, Cronbach’s alpha, descriptive statistics and the Kruskal–Wallis test. Hypotheses were validated through partial least squares structural equation modelling (PLS-SEM). Findings The study revealed that out of the 43 identified barriers, inadequate system integration, challenges in guaranteeing interoperability, university education on the subject is deficient, and new system compatibility with legacy systems are the main barriers to implementing DT for sustainable construction practices in the AEC Industry of Nigeria. Research limitations/implications The study was conducted in Nigeria with a focus on the Federal Capital Territory. The study identified the barriers of DT in the construction sector. Practical implications This study developed and assessed a theoretical framework, examining the relationships between variables. The findings have important implications for the construction industry, offering opportunities to improve construction processes. Furthermore, the study will help improve sustainable practices within the built environment. Originality/value The study categorised the barriers of DT into the following: system integration; security-related; performance-related; organizational-related; data quality issues and environmental related issues.