건설 분야에서의 생성적 적대 신경망 (GAN) 적용 연구 동향 분석

Abstract

This study analyzes the potential applications and challenges of employing Generative Adversar- ial Networks (GAN) in the Architecture, Engineering, and Construction (AEC) industry, which has recently gained significant attention. The study explains the basic structure and evolution of GAN, providing a comprehensive understanding of the concept. Using the PRISMA methodol- ogy, 36 relevant papers were systematically reviewed to assess the current application of GAN in the AEC industry. The results indicate that GAN can be effectively utilized in various areas such as construction material performance prediction, architectural design automation, construc- tion site monitoring, and safety management. However, several challenges, including technical complexity, cost, and reliability issues, remain. This study highlights the potential of GAN to become an innovative tool within the AEC industry.