Skip to content

Automated daily report generation from construction videos using ChatGPT and computer vision

Xiao, B., Wang, Y., Zhang, Y., Chen, C., & Darko, A. (2024). Automated daily report generation from construction videos using ChatGPT and computer vision. Automation in Construction, 168, 105874-. https://doi.org/10.1016/j.autcon.2024.105874

View Publication

Abstract

Daily reports are important in construction management, informing project teams about status, enabling timely resolutions of delays and budget issues, and serving as official records for disputes and litigation. However, current practices are manual and time-consuming, requiring engineers to physically visit sites for observations. To fill this gap, this paper proposes an automated framework to generate daily construction reports from on-site videos by integrating ChatGPT and computer vision (CV)-based methods. The framework utilizes CV methods to analyze video footage and extract relevant productivity and activity information, which is then fed into ChatGPT using proper prompts to generate daily reports. A web application is developed to implement and validate the framework on a real construction site in Hong Kong, generating daily reports over a month. This research enhances construction management by significantly reducing documentation efforts through generative artificial intelligence, with potential applications in jobsite safety management, quality reporting, and stakeholder communication.

Keywords

Construction daily report generation; Computer vision; ChatGPT; Construction management; Project documentation

Online toolkits for collaborative and inclusive global research in urban evolutionary ecology

Savage, A. M., Willmott, M. J., Moreno‐García, P., Jagiello, Z., Li, D., Malesis, A., Miles, L. S., Román‐Palacios, C., Salazar‐Valenzuela, D., Verrelli, B. C., Winchell, K. M., Alberti, M., Bonilla‐Bedoya, S., Carlen, E., Falvey, C., Johnson, L., Martin, E., Kuzyo, H., Marzluff, J., … Gotanda, K. M. (2024). Online toolkits for collaborative and inclusive global research in urban evolutionary ecology. Ecology and Evolution, 14(6), e11633-n/a. https://doi.org/10.1002/ece3.11633

View Publication

Abstract

Urban evolutionary ecology is inherently interdisciplinary. Moreover, it is a field with global significance. However, bringing researchers and resources together across fields and countries is challenging. Therefore, an online collaborative research hub, where common methods and best practices are shared among scientists from diverse geographic, ethnic, and career backgrounds would make research focused on urban evolutionary ecology more inclusive. Here, we describe a freely available online research hub for toolkits that facilitate global research in urban evolutionary ecology. We provide rationales and descriptions of toolkits for: (1) decolonizing urban evolutionary ecology; (2) identifying and fostering international collaborative partnerships; (3) common methods and freely-available datasets for trait mapping across cities; (4) common methods and freely-available datasets for cross-city evolutionary ecology experiments; and (5) best practices and freely available resources for public outreach and communication of research findings in urban evolutionary ecology. We outline how the toolkits can be accessed, archived, and modified over time in order to sustain long-term global research that will advance our understanding of urban evolutionary ecology.

Driving factors for the adoption of green finance in green building for sustainable development in developing countries: The case of Ghana

Debrah, C., Chan, A. P. C.Darko, A.Ries, R. J.Ohene, E., & Tetteh, M. O. (2024). Driving factors for the adoption of green finance in green building for sustainable development in developing countries: The case of GhanaSustainable Development, https://doi.org/10.1002/sd.3022

View Publication

Abstract

While there are many motivating factors for green finance (GF) implementation, a comprehensive taxonomy of these variables is lacking in the literature, especially for green buildings (GBs). This study aims to analyze the criticality and interdependence of GF-in-GB's driving factors. This study develops a valid set of factors to justify the interrelationships among the drivers. The drivers of GF-in-GB are qualitative in nature, and uncertainties exist among them due to linguistic preferences. This study applies the fuzzy Delphi method to validate eight drivers under uncertainties. Fuzzy Decision-Making Trial and Evaluation Laboratory (FDEMATEL) with qualitative information is used to determine the interrelationships among the drivers. The drivers were grouped under two categories: prominent drivers and cause-effect drivers. The findings revealed that “increased awareness of GF models in GB” and “preferential capital requirements for low-carbon assets” are the top two most prominent/important drivers of GF-in-GB. In Ghana, the top three cause group drivers are “climate commitment,” “improved access to and lower cost of capital,” and “favorable macroeconomic conditions and investment returns.” Drivers with the highest prominence values have the potential to affect and/or be affected by other drivers; therefore, managers and policymakers should prioritize promoting or pursuing these drivers in the short term. On the other hand, it is important to pay more than equal attention to the drivers with the highest net cause values because they have the largest long-term impact on the entire system. The theoretical and practical implications of the study are discussed, enhancing understanding and decision-making in GF-in-GB.

Challenges to energy retrofitting of existing office buildings in high-rise high-density cities: The case of Hong Kong

Linyan Chen, Amos Darko, Mayowa I. Adegoriola, Albert P.C. Chan, Yang Yang, Mershack O. Tetteh, “Challenges to energy retrofitting of existing office buildings in high-rise high-density cities: The case of Hong Kong,” Energy and Buildings, Volume 312, 2024, 114220, ISSN 0378-7788, https://doi.org/10.1016/j.enbuild.2024.114220.

View Publication

Abstract

Achieving carbon neutrality by 2050 has become a global goal, sparking concerns regarding energy consumption and carbon emissions in building operations. Office buildings in high-rise high-density cities serve as central business districts, contributing significantly to the city’s economic activity and consuming a lot of energy. The process of retrofitting existing office buildings for energy efficiency in high-rise high-density cities tends to be challenging. However, there is a lack of comprehensive understanding of the challenges involved in office buildings’ energy retrofitting, as they have not been thoroughly explored. This study aims to investigate the challenges to the existing office building energy retrofitting (EOBER) in high-rise high-density cities with real cases in Hong Kong. Initially, a systematic literature review was undertaken to identify 49 potential EOBER challenges and categorized into seven groups: technical, financial, institutional, social, environmental, regulatory, and other categories. Afterward, 23 EOBER challenges were identified through 24 semi-structured interviews with 36 real office building energy retrofitting cases in Hong Kong. Moreover, these challenges were quantified by the Z-numbers-based Delphi survey and analysis. Results show that regulatory challenges are the primary challenges, followed by financial challenges. The lack of government incentives, policies, legislation and regulations significantly hinders practitioners’ ability to engage in energy retrofitting initiatives. The long payback period of building energy retrofitting poses a critical financial concern for practitioners embracing such initiatives. In the end, this research proposed integrated strategies to tackle these challenges and increase building energy efficiency, including launching financial and regulatory incentives, shortening the interval for mandatory energy audits, disseminating knowledge, and diversifying finance channels of building energy retrofitting. The findings contribute to the body of knowledge by employing systems thinking to identify and evaluate EOBER challenges in high-rise high-density cities through empirical methodologies. Moreover, this study provides valuable references for practitioners in navigating these challenges and minimizing risks associated with the retrofitting process.

Rebecca Bachman

Rebecca’s long-term design and research interests involve exploring a gap between built environment design fields and public health research. She believes that an essential component of design activism must be research that can inform equitable policy. Rebecca holds a Master of Landscape Architecture and Graduate Certificate of Global Health from the University of Washington, and is a current NIH Fogarty Scholar. She has spent recent years working with the design-action-research firm Traction on design and research projects that examine built environments as social determinants of human and ecological health in marginalized urban communities in Peru. In addition to English, Rebecca speaks Spanish and Portuguese, and has a blended background in international relations, journalism, and horticulture—all of which is integral to her current interdisciplinary work.

Automating building environmental assessment: A systematic review and future research directions

T.A.D.K. Jayasanka, Amos Darko, D.J. Edwards, Albert P.C. Chan, Farzad Jalaei, Automating building environmental assessment: A systematic review and future research directions, Environmental Impact Assessment Review, Volume 106, 2024, 107465, ISSN 0195-9255, https://doi.org/10.1016/j.eiar.2024.107465.

View Publication

Abstract

Building environmental assessment (BEA) is critical to improving sustainability. However, the BEA process is inefficient, costly, and often inaccurate. Because automation has the potential to enhance the efficiency and accuracy of the BEA process, studies have focused on automating BEA (ABEA). Updated until now, a comprehensive analysis of prevailing literature on ABEA remains absent. This study conducts the first comprehensive systematic analysis appraising the state-of-the-art of research on ABEA. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guided to systematically analyse 91 relevant studies. Results uncover that only 29.7% of BEA systems worldwide have automated their processes, with the US LEED residing at the vanguard of automation efforts. The New Buildings scheme was mostly focused on, while largely ignoring other schemes, e.g., Existing Buildings. Five key digital approaches to ABEA were revealed, namely building information modelling (BIM) and plug-in software, BIM-ontology, data mining and machine learning, cloud-BIM, and digital twin-based approaches. Based on identified gaps, future research directions are proposed, specifically: using data mining and machine learning models for ABEA; development of a holistic cloud-based approach for real-time BEA; and digital twin for dynamic BEA. This study generates a deeper understanding of ABEA and its theoretical implications, such as major constructs and emerging perspectives, constitute a basis for holistic, and innovation in, BEA.

Change Stories project begins 4-day residential event in Bogotá, Colombia

The Change Stories project is an research project with collaboration from academics and their community-based partners in Belfast (Queen’s University), Northern Ireland, Belo Horizonte (Federal University of Minas Gerais Brazil and Observatory for Urban Health), Brazil, and Bogotá (Universidad de los Andes), Colombia. Additional collaborators include USA and Internationally-based advisory group members, who work within their communities. The project is funded by the Robert Wood Johnson Foundation. The 3 case study cities in the Change Stories project are Belfast, Northern…

Magdalena Haakenstad

Magdalena Haakenstad, Postdoctoral Scholar in Urban Planning for Health in the University of Washington Department of Urban Design and Planning in Seattle, Washington, USA. She is a cultural anthropologist interested in environmental health, means and strategies of political negotiation, visual communication in public space, and decolonizing methodologies. As a part of her research, she had an opportunity to work with historically marginalized communities in Mexico, the US and Slovakia on public art projects, storytelling, photo essays, and filmmaking to help amplify their voices. She holds a PhD in General Anthropology from Charles University in Prague, Czech Republic.