Journal of
the Architectural Institute of Korea

  • ISSN : 2733-6239(Print)
  • ISSN : 2733-6247(Online)

Architectural Institute of Korea (AIK) members can check the full-text content of the published paper by logging in.

Everyone can download published papers by clicking (DBpia, a Korean academic information portal) and paying a download fee.

Validating an AI Model for Building Lifespan Prediction Using Big Data

지석원(Ji, Suk-Won)

Accurately estimating a building's lifespan is crucial for assessing its asset value and determining its economic and environmental feasibility, which is key for decision-making in the construction industry. However, because it's nearly impossible to precisely estimate the lifespan of each building due to the wide range of influencing factors, most studies have used uniform lifespans based on the building's primary structural type. To address this limitation, 1,812,700 records were analyzed of buildings constructed and demolished in Korea to predict each building's lifespan with greater accuracy. Based on the previous study, a prediction model was developed using both deep learning and traditional machine learning methods. This study evaluated whether the building lifespan prediction model experienced overfitting based on the data period used to create the model. A performance evaluation was also conducted, comparing models using only key factors to those using a broader set of factors. The results showed that among the machine learning models, the artificial neural network model, a nonlinear approach, maintained high predictive accuracy without overfitting, regardless of the data period used. The model that used all available factors performed better than those based on just a few key factors. This research demonstrates the viability of using big data and AI for building lifespan prediction, providing a more reliable method for estimating building lifespan tailored to each building's unique characteristics. This approach meets a growing societal demand for more accurate building lifespan predictions.

Identifying Risk Factors at Various PF Business Stages and Investigating Risk Management Measures

이윤홍(Lee, YunHong) ; 서원경(Seo, Wonkyoung)

The real estate market faced a downturn with declining sales, exacerbated by increased construction costs due to conflicts in Russia and Ukraine. Additionally, interest rates on PF loans surged following the Bank of Korea's base interest rate hike. Given these challenges, ongoing real estate development lacks feasibility as sales dwindle and business expenses soar. These risks extend beyond development firms, affecting mid-sized construction companies, second-tier financial institutions, and trust companies, posing broader risks to society. To mitigate these risks, PF risk factors were identified through credible literature and expert opinions. Using ANOVA variance analysis, key PF factors considered by each organization were examined and high-risk items were pinpointed by assessing the correlation between risk factors.

A Study on the Problem Derivation and Improvement Direction through the Analysis of the Current Status of Building Management Act

박찬영(Park, Chan Young)

The Building Management Act, enacted on May 1, 2020, emerged in response to the escalating challenges posed by unmanaged and aging buildings, prompting a separation from the Building Code. The legislation sought to institute comprehensive building management plans, addressing the entire life cycle of a building, including maintenance, safety inspections, and post-construction energy conservation. This study, conducted three years post-enforcement, scrutinized the efficacy of the recently introduced building management plan. Findings unveiled that approximately 40% of plans were deemed unsatisfactory, necessitating thorough review. Key contributors to this low fidelity encompassed the absence of system establishment, fostering a limited understanding between builders and officials. Furthermore, issues such as the system's unclear establishment points and the submission of handwritten, non-digitized PDF documents posed substantial hurdles for verification and future utility. To rectify these challenges, proposed solutions include educating authors and government officials, implementing an active follow-up system, and enhancing the system. This study marks the inaugural investigation into the status of building management plans under the Building Management Act, aiming to pinpoint problems and propose effective improvement strategies. The aspiration is that these measures will foster the development and robust establishment of an efficient system.

Analyzing Campus Spatial Arrangements to Support Multidisciplinary Collaboration Among Convergence Majors

이소영(Lee, So Young) ; 이명아(Lee, Myung-A)

Many universities aim to collaborate on education, research, and knowledge valorization, making it crucial to create physical spaces that encourage interdisciplinary work. The purpose of this study is to examine the convergence majors offered at the university. It aims to identify which majors emphasize convergence education, to understand the layout and spatial characteristics of the campus, and to see how convergence-focused majors are distributed across the campus. Additionally, it explores the proximity and spatial arrangement of the key buildings related to these majors. A network analysis of convergence majors was performed using the NodeXL program. Once the convergence majors were defined, the spatial characteristics among them were explored with the Depthmap X program. Axial analysis focused on connectivity, integration, choice and mean depth in various building areas across the campus. The administration building, student lounge, and library were all centrally located on campus. Among key convergence majors such as the medical, engineering, and business schools, the business school tended to be at the campus's center, close to other major buildings. Given the campus layout, the College of Business Administration building is a flexible and collaborative space, conveniently located near central areas and other important buildings.

Exploring the Impact of AI-based Prompt Design Method on Architectural Design Process and Designers’ Role

창쩌위안(Chang, Ze-Yuan) ; 한정원(Han, Jeong-won)

This study explores AI-based prompt design methods and their impact on architectural design process and designers, examining how these approaches alter traditional design processes and redefine architects' roles. By synthesizing the insights of six design practitioners, this study contrasts prompt design with conventional methods, exploring their practical applications and potential future directions. It also delves into the challenges and limitations posed by AI technologies in this evolving context. This study reveals that AI-based prompt design represents a new design paradigm, where prompted keywords play a central role in streamlining workflow, boosting creativity, and facilitating the translation of concepts into architectural schematics. This paradigm shift transforms architects from solitary designers into AI collaborators, necessitating a broader skill set in design literacy. Prompt design, driven by designer-suggested keywords or prompts, enables a more direct and precise process, allowing architects to clearly define their goals and scope. The evolving role of architects reflects this collaborative approach, with AI handling visualization tasks while designers focus on design decision-making. This shift underscores the ongoing relevance and expertise of architects, highlighting their new role as diversified engineering designers in a dynamic AI-driven environment.

Development and Application of an Architectural Component Combination Algorithm for Generative Design-based DfMA Design Support

홍순민(Hong, Soon Min) ; 김근재(Kim, Geunjae) ; 구형모(Gu, Hyeongmo) ; 추승연(Choo, Seungyeon)

The AEC industry faces persistent challenges such as an aging workforce and low productivity. Off-Site Construction (OSC) and Design for Manufacture and Assembly (DfMA) are emerging as solutions to these issues, receiving significant attention and application in the United States, Singapore, Hong Kong, and recently in Korea. DfMA, a design method that incorporates front loading at the design stage, involves analyzing and disassembling all aspects of a building to design modules optimized for transport and installation. Recognized for its potential to enhance safety, reduce construction time, and increase productivity, DfMA addresses the limitations of OSC construction, particularly errors in the complex design and fabrication stages. To enhance OSC's effectiveness and productivity, this study proposes the development of a BIM-based generative design algorithm. This algorithm aims to rapidly generate numerous combinations of building components using domestic OSC materials, applying DfMA principles at the design stage. The research methodology includes a review of DfMA concepts, analysis of case studies and prior research, investigation and classification of domestic OSC building components, and development of a generative design algorithm for building component combinations. By applying this algorithm, the study seeks to identify and verify the optimal combination of components, considering budget and selected types. Despite DfMA's advantages, issues such as design complexity, uniformity, and assembly errors have been extensively studied. The algorithm introduced in this research could contribute significantly to DfMA-based architectural design, improving fabrication and assembly efficiency early in the design process.

A Study on The Characteristics of Parking Demand for Factory Facilities in Industrial Complex Considering Company Size and Manufacturing Industry

양승길(Yang, Seung-kil) ; 양승우(Yang, Seung-woo)

The study purpose to analyze parking demand characteristics based on industry type and company size by assessing parking availability within and outside factory facilities in industrial zones. Manufacturing industry averages 238.8㎡/car for parking space. Parking demand correlates with factory floor area, notably low for machine and electronics industries with less than 2,000㎡ and 2,000-3,000㎡ respectively. However, the metal industry with 5,000-10,000㎡ floor area shows a 118% increase in parking space per unit. Parking demand varies by company size, with larger floor areas leading to increased parking needs. Small-scale enterprises exhibit a 63% increase, small enterprises 69%, and medium-sized enterprises 75% increase compared to current standards, indicating differing parking demands across company sizes. For companies with less than 2,000㎡ floor areas across industries, parking space per unit is at 47% of the current standard. Electronics and chemical industries consistently show lower parking space per unit across company sizes. Conversely, small-scale enterprises in metal and machinery industries (3,000-5,000㎡) and small enterprises in the metal industry (5,000-10,000㎡) exceed current standards for parking space per unit. This analysis underscores the need for tailored parking solutions based on industry and company size within industrial complexes.

Analyzing Utilization Strategies and Factors of Closed School Occurrence

서형경(Seo, Hyung-Kyung) ; 전영훈(Jeon, Young-Hoon)

Closed schools present a significant challenge as they are often neglected and disconnected from their local communities, constituting 65.6% of the 3,896 closed schools. This abandonment not only entails the loss of a valuable facilities but also eradicates collective memory and historical significance associated with these institutions, thereby impacting the local living environment. Consequently, this study aims to comprehensively analyze the issues surrounding the utilization of closed schools, examining their current state and proposing effective strategies for improvement. Drawing from the background of closed school occurrences, three regions anticipating an increase in closed schools and facing social issues were selected for analysis. The examination of closed school usage progress in three aspects: the process of practical use, utilization patters, and policy and financial assistance. Insights from the practices in foreign countries underscore the importance of guidelines for repurposing closed schools, obtaining permissions for remodeling, exploring diverse utilization opportunities, and fostering collaborations with organizations beyond the Office of Education. By implementing these findings, closed schools can potentially transform into significant spaces within their communities, enriching residents’ lives and serving as multifunctional community hubs.

Examining Trends in Model House Research From 1994 to 2023

정수환(Cheong, Soo-Hwan) ; 백진(Baek, Jin)

Since the Korea National Housing Corporation (KNHC) constructed the first model house in 1969 to experiment with the pre-sale system for apartment housing, this concept has attracted attention across various disciplines in Korea. However, a comprehensive analysis of trends in model-house research has been lacking until now. This study aims to explore the trends of model houses through a systematic literature review. A total of 95 relevant papers published in Korean academic journals from 1994 to 2023 were examined, utilizing quantitative analysis of sources searched from databases like RISS, DBPIA, and AURIC. The research was then categorized into four groups using content and frequency analysis based on the perception of model houses: apartment interior units, independent facilities, virtual technology, and others such as concepts reflecting housing sales, cultural, philosophical, and urban phenomena of Korea. The frequency analysis revealed that the majority of research focuses on apartment interior units at 45.3%, with significant contributions from the other groups at 20.0%, 21.3%, and 13.7%, respectively. These results not only deepen our understanding of model house research but also substantially broaden perspectives, providing a solid foundation for future studies and enhancing the efficiency of literature reviews for new researchers. This is particularly useful for distinguishing between prominently addressed and less emphasized topics in model house research.

Le Corbusier’s Early Perspectives on Urbanism in La Construction des Villes, Focused on the Topics of Blocks, Streets and Squares

김영민(Kim, Youngmin)

This study explores Le Corbusier's early urban planning ideas as presented in his unpublished work, La Construction des Villes 1910-1911. It contradicts the common view that Le Corbusier was exclusively a proponent of rigid geometric city designs during the 1920s. Instead, La Construction des Villes indicates that he was influenced by Camillo Sitte's more organic approach to city planning. This stands in contrast to Le Corbusier's later and better-known geometric vision, the Radiant City. While discussing his ideal urban block, Le Corbusier shows a preference for layouts with landscaped courtyards, creating safe, green areas for residents. The study also examines his views on streets and squares. Although he often promoted curved streets for their visual appeal and ability to navigate complex terrain, he also recognized the practicality of straight streets in major cities like Paris. Regarding public squares, Le Corbusier shared Sitte's criticism of 19th-century designs but still saw value in certain geometric Baroque squares depending on the context. A key concept in Le Corbusier's approach is la corporalite, which focuses on creating outdoor spaces that function like rooms, blending aesthetics with utility. Through this analysis of La Construction des Villes, this study offers fresh insights into Le Corbusier's urban planning philosophy, challenging conventional interpretations and encouraging a reexamination of his approach to the relationship between city planning and architecture.