Socrates Yiannoudes
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Latest Publications
- 22 September 2026
- Published by SAGE Publications in International Journal of Architectural Computing
Practice readiness in generative and AI-driven architectural design: A systematic assessment
Generative and AI-driven design methods are proliferating in architectural research. Yet, their potential impact on day-to-day practice remains unclear. This review maps 67 case studies drawn from 65 peer-reviewed publications, across nine generative categories organised under four computational approaches; rule-driven, search-based, data-driven and hybrid. Each case is assessed on a three-tier scale through a Practice Engagement Index (PEI) comprising six indicators: workflow accessibility, real-world constraints alignment, common building type, geometric viability, funding, and physical construction application. Results show that most case studies engage convincingly with realistic constraints (62.6%) and common building types (58.2%), yielding potentially buildable geometry (47.7%). Far fewer achieve sufficient workflow accessibility (17.9% high tier, with 50.7% scoring low), secure industry funding or reach physical construction. Although publication volume has surged since 2020, two-thirds of studies cluster at moderate readiness (PEI ≥ 1.0) and very few approach high engagement (PEI > 1.5). These results suggest that researchers increasingly acknowledge professional realities yet still struggle to translate technical novelty into routinised tools and construction-phase implementation. By pairing a taxonomy of generative categories with a replicable, indicator-based assessment index, this study offers a transparent instrument for comparing practice-readiness across methods and for tracking progress over time. Further progress will depend less on algorithmic sophistication than on containerized plug-ins for mainstream CAD/BIM packages and IFC pipelines, industry partnership, and pathways to construction.
- 22 September 2026
- Published by SAGE Publications in International Journal of Architectural Computing
Practice readiness in generative and AI-driven architectural design: A systematic assessment
Generative and AI-driven design methods are proliferating in architectural research. Yet, their potential impact on day-to-day practice remains unclear. This review maps 67 case studies drawn from 65 peer-reviewed publications, across nine generative categories organised under four computational approaches; rule-driven, search-based, data-driven and hybrid. Each case is assessed on a three-tier scale through a Practice Engagement Index (PEI) comprising six indicators: workflow accessibility, real-world constraints alignment, common building type, geometric viability, funding, and physical construction application. Results show that most case studies engage convincingly with realistic constraints (62.6%) and common building types (58.2%), yielding potentially buildable geometry (47.7%). Far fewer achieve sufficient workflow accessibility (17.9% high tier, with 50.7% scoring low), secure industry funding or reach physical construction. Although publication volume has surged since 2020, two-thirds of studies cluster at moderate readiness (PEI ≥ 1.0) and very few approach high engagement (PEI > 1.5). These results suggest that researchers increasingly acknowledge professional realities yet still struggle to translate technical novelty into routinised tools and construction-phase implementation. By pairing a taxonomy of generative categories with a replicable, indicator-based assessment index, this study offers a transparent instrument for comparing practice-readiness across methods and for tracking progress over time. Further progress will depend less on algorithmic sophistication than on containerized plug-ins for mainstream CAD/BIM packages and IFC pipelines, industry partnership, and pathways to construction.
- Open Access
- 10 October 2025
- Published by MDPI AG in Architecture
- Vol. 5 (4) , 94
Shaping Architecture with Generative Artificial Intelligence: Deep Learning Models in Architectural Design Workflow
Deep-learning generative AI promises to transform architectural design, yet its potential employment and ready-to-use capacity for professional workflows are unclear. This study presents a systematic review conducted in accordance with PRISMA 2020 guidelines, synthesizing peer-reviewed work from 2015 to 2025 to assess how GenAI methods align with architectural practice. A total of 1566 records were initially retrieved across databases, of which 42 studies met eligibility criteria after structured screening and selection. Each was evaluated using five indicators with a three-tier rubric: Output Representation Type, Pipeline Integration, Workflow Standardization, Tool Readiness, and Technical Skillset. Results show that most outputs are raster images or non-editable objects, with only a minority producing CAD/BIM-ready geometry. Workflow pipelines are often fragmented with manual hand-offs and most GenAI methods map only onto the early conceptual design stage. Prototypes frequently require bespoke coding and advanced expertise. These findings indicate a persistent gap between experimentation with ideation-oriented GenAI and the pragmatism of CAD/BIM-centered delivery. By framing the proposed rubric as a workflow maturity model, this review contributes a replicable benchmark for assessing practice readiness and identifying pathways toward mainstream adoption. For GenAI to move from prototypes to mainstream architectural design practice, it is essential to address not only technical barriers, but also cultural issues such as professional skepticism and reliability concerns, as well as ecosystem challenges of data sharing, authorship, and liability.
- 10 October 2025
- Published by MDPI AG in Architecture
- Vol. 5 (4) , 94
Open AccessShaping Architecture with Generative Artificial Intelligence: Deep Learning Models in Architectural Design Workflow
Deep-learning generative AI promises to transform architectural design, yet its potential employment and ready-to-use capacity for professional workflows are unclear. This study presents a systematic review conducted in accordance with PRISMA 2020 guidelines, synthesizing peer-reviewed work from 2015 to 2025 to assess how GenAI methods align with architectural practice. A total of 1566 records were initially retrieved across databases, of which 42 studies met eligibility criteria after structured screening and selection. Each was evaluated using five indicators with a three-tier rubric: Output Representation Type, Pipeline Integration, Workflow Standardization, Tool Readiness, and Technical Skillset. Results show that most outputs are raster images or non-editable objects, with only a minority producing CAD/BIM-ready geometry. Workflow pipelines are often fragmented with manual hand-offs and most GenAI methods map only onto the early conceptual design stage. Prototypes frequently require bespoke coding and advanced expertise. These findings indicate a persistent gap between experimentation with ideation-oriented GenAI and the pragmatism of CAD/BIM-centered delivery. By framing the proposed rubric as a workflow maturity model, this review contributes a replicable benchmark for assessing practice readiness and identifying pathways toward mainstream adoption. For GenAI to move from prototypes to mainstream architectural design practice, it is essential to address not only technical barriers, but also cultural issues such as professional skepticism and reliability concerns, as well as ecosystem challenges of data sharing, authorship, and liability.
- Open Access
- 1 September 2025
- Published by MDPI AG in Preprints.org
Shaping Architecture with Generative Artificial Intelligence: Deep Learning Models in Architectural Design Workflow
Deep-learning generative AI promises to transform architectural design, yet its potential employment and ready-to-use capacity for everyday workflows are unclear. This study systematically reviews peer-reviewed work from 2015–2025 to assess how GenAI methods align with architectural practice. Following database searches and subject-area filtering, 42 studies were included from 1,566 records. Each was evaluated with a five-indicator, three-tier rubric: Output Representation Type (ORT), Pipeline Integration (PI), Workflow Standardization (WS), Tool Readiness (TR), and Technical Skillset (TS). Results show outputs are concentrated in non-native formats (≈40% raster imagery; ≈45% meshes/voxels/graphs), with relatively few CAD/BIM-native results (≈15%). Toolchains are often fragmented (PI: ≈43% Tier-0 with ≥4 steps; ≈40% Tier-1 with 2–3 tools; ≈17% Tier-2 single-platform). Most studies map to schematic-only design stage (WS Tier-1 ≈69%), and multi-stage, CAD/BIM-compatible pipelines remain uncommon (WS Tier-2 ≈12%). Prototypes frequently require bespoke coding (TR Tier-0 ≈65%) and advanced expertise (TS Tier-0 ≈74%). These findings indicate a persistent gap between experimentation with ideation-oriented GenAI and the pragmatism of CAD/BIM-centered delivery. Advancing practice readiness will require native CAD/BIM outputs, tighter plug-in/API integration, tools that bridge heterogeneous file formats and metadata export, and packaging ML modules into CAD/BIM environments that lowers skill demands. Limitations include the academic focus of the corpus and rapid field evolution.
- 1 September 2025
- Published by MDPI AG in Preprints.org
Open AccessShaping Architecture with Generative Artificial Intelligence: Deep Learning Models in Architectural Design Workflow
Deep-learning generative AI promises to transform architectural design, yet its potential employment and ready-to-use capacity for everyday workflows are unclear. This study systematically reviews peer-reviewed work from 2015–2025 to assess how GenAI methods align with architectural practice. Following database searches and subject-area filtering, 42 studies were included from 1,566 records. Each was evaluated with a five-indicator, three-tier rubric: Output Representation Type (ORT), Pipeline Integration (PI), Workflow Standardization (WS), Tool Readiness (TR), and Technical Skillset (TS). Results show outputs are concentrated in non-native formats (≈40% raster imagery; ≈45% meshes/voxels/graphs), with relatively few CAD/BIM-native results (≈15%). Toolchains are often fragmented (PI: ≈43% Tier-0 with ≥4 steps; ≈40% Tier-1 with 2–3 tools; ≈17% Tier-2 single-platform). Most studies map to schematic-only design stage (WS Tier-1 ≈69%), and multi-stage, CAD/BIM-compatible pipelines remain uncommon (WS Tier-2 ≈12%). Prototypes frequently require bespoke coding (TR Tier-0 ≈65%) and advanced expertise (TS Tier-0 ≈74%). These findings indicate a persistent gap between experimentation with ideation-oriented GenAI and the pragmatism of CAD/BIM-centered delivery. Advancing practice readiness will require native CAD/BIM outputs, tighter plug-in/API integration, tools that bridge heterogeneous file formats and metadata export, and packaging ML modules into CAD/BIM environments that lowers skill demands. Limitations include the academic focus of the corpus and rapid field evolution.
- Open Access
- 16 August 2025
- Published by Universitas Atma Jaya Yogyakarta in Journal of Artificial Intelligence in Architecture
- Vol. 4 (2) , 70-88
Text-to-Image Generative Αrtificial Intelligence As Conceptual Mechanism For Architectural Design
This study investigates the potential of text-to-image generative AI applications to be used as exploration mechanisms for architectural design rather than mere visualisation tools. We examine whether text-to-image generative AI applications can "understand" architecture's spatial and geometric configuration in meaningful and creative ways. Our experimental method used a series of architectural plans and artistic patterns as image prompts, combined with varied text prompts and parameters to evaluate Midjourney's capacity for geometric and spatial interpretation and creativity. The experiments focused on the application's capacity to (1) generate diagrams from architectural plans, (2) create variations of two-dimensional drawings and patterns, and (3) transform two-dimensional layouts into three-dimensional spaces. Results indicate that Midjourney can generate meaningful abstractions and spatial configurations but struggles with strict spatial fidelity, particularly for asymmetrical or fragmented layouts. The study highlights the significance of short prompts and increased image weight parameters for improving fidelity and structural coherence. While Midjourney exhibits potential as a conceptual tool for highly creative architectural exploration, it lacks the precision required for deterministic spatial manipulation. Dall-E performs significantly better in fidelity and alignment to the original drawing, but further research is needed to determine its potential use as an assistive tool for architectural design. The findings contribute to the discourse on AI-assisted architectural design for creative ideation and suggest further research into the design capacities of other similar models.
- 16 August 2025
- Published by Universitas Atma Jaya Yogyakarta in Journal of Artificial Intelligence in Architecture
- Vol. 4 (2) , 70-88
Open AccessText-to-Image Generative Αrtificial Intelligence As Conceptual Mechanism For Architectural Design
This study investigates the potential of text-to-image generative AI applications to be used as exploration mechanisms for architectural design rather than mere visualisation tools. We examine whether text-to-image generative AI applications can "understand" architecture's spatial and geometric configuration in meaningful and creative ways. Our experimental method used a series of architectural plans and artistic patterns as image prompts, combined with varied text prompts and parameters to evaluate Midjourney's capacity for geometric and spatial interpretation and creativity. The experiments focused on the application's capacity to (1) generate diagrams from architectural plans, (2) create variations of two-dimensional drawings and patterns, and (3) transform two-dimensional layouts into three-dimensional spaces. Results indicate that Midjourney can generate meaningful abstractions and spatial configurations but struggles with strict spatial fidelity, particularly for asymmetrical or fragmented layouts. The study highlights the significance of short prompts and increased image weight parameters for improving fidelity and structural coherence. While Midjourney exhibits potential as a conceptual tool for highly creative architectural exploration, it lacks the precision required for deterministic spatial manipulation. Dall-E performs significantly better in fidelity and alignment to the original drawing, but further research is needed to determine its potential use as an assistive tool for architectural design. The findings contribute to the discourse on AI-assisted architectural design for creative ideation and suggest further research into the design capacities of other similar models.