Search results

This study aims to present a twofold machine learning (ML) model, namely, EDU-AI, and its implementation in educational buildings. The specific focus is on classroom layout design, which is investigated regarding implementation of ML in the early phases of design.

This study introduces the framework of the EDU-AI, which adopts generative adversarial networks (GAN) architecture and Pix2Pix method. The processes of data collection, data set preparation, training, validation and evaluation for the proposed model are presented. The ML model is trained over two coupled data sets of classroom layouts extracted from a typical school project database of the Ministry of National Education of the Republic of Turkey and validated with foreign classroom boundaries. The generated classroom layouts are objectively evaluated through the structural similarity method (SSIM).

The implementation of EDU-AI generates classroom layouts despite the use of a small data set. Objective evaluations show that EDU-AI can provide satisfactory outputs for given classroom boundaries regardless of shape complexity (reserved for validation and newly synthesized).

EDU-AI specifically contributes to the automation of classroom layout generation using ML-based algorithms. EDU-AI’s two-step framework enables the generation of zoning for any given classroom boundary and furnishing for the previously generated zone. EDU-AI can also be used in the early design phase of school projects in other countries. It can be adapted to the architectural typologies involving footprint, zoning and furnishing relations.

To date, there are no studies in the literature that provide a comprehensive understanding of the interrelationships between the slenderness ratio and the main design criteria in supertall towers (=300 m). In this paper, this important issue was explored using detailed data collected from 75 cases.

This paper was carried out with a comprehensive literature review including the database of the Council on Tall Buildings and Urban Habitat(CTBUH) (CTBUH, 2022), peer-reviewed journals, MSc theses and PhD dissertations, conference proceedings, fact sheets, architectural and structural magazines and other Internet sources. In this study, the case study method was also used to gather and consolidate information about supertall towers to analyze the interrelationships. Cases were 75 supertall buildings in various countries [44 from Asia (37 from China), 16 from the Middle East (6 from Dubai, the United Arab Emirates), 11 from the United States of America and 3 from Russia, 1 from the UK].

The paper's findings highlighted as follows: (1) for buildings in the height range of 300–399 m, the slenderness ratio was usually between 7 and 7.9 and megatall towers were frequently built at a slenderness ratio of 10–15; (2) the median slenderness ratio of buildings in the 400–599 m height ranges was around 8.6; (3) a trend towards supertall slender buildings (=8) was observed in Asia, the Middle East and North America; (4) residential, office and mixed-use towers had a median slenderness ratio of over 7.5; (5) all building forms were utilized in the construction of slender towers (>8); (6) the medium slenderness ratio was around 8 for supertall buildings constructed with outriggered frame and tube systems; (7) especially concrete towers reached values pushing the limits of slenderness (>10) and (8) since the number of some supertall building groups (e.g. steel towers) was not sufficient, establishing a scientific relationship between aspect ratio and related design criteria was not possible.

To date, there are no studies in the literature that provide a comprehensive understanding of the interrelationships between the slenderness ratio and the main design criteria in supertall towers (=300 m). This important issue was explored using detailed data collected from 75 cases.