Journal of Umm Al-Qura University for Engineering and Architecture

https://doi.org/10.1007/s43995-025-00186-1

REVIEW

The role of artificial intelligence (AI) in architectural design: a

systematic review of emerging technologies and applications
Ibraheem Nooruddeen Albukhari1

Received: 6 April 2025 / Accepted: 13 July 2025

© The Author(s) 2025

Abstract
Recent advancements in Artificial Intelligence (AI) are reshaping architectural design by enhancing key design tech-
niques such as spatial planning, parametric modeling, generative design, and performance-based analysis. AI methods like
machine learning (ML) and predictive modeling support tasks from material selection to structural optimization, improv-
ing efficiency, sustainability, and creativity. By analyzing large datasets and automating complex processes, AI empowers
architects to explore innovative solutions responsive to user needs and environmental factors. This study systematically
reviews how AI plays a part in the architectural workflow, focusing on its ability to enhance creativity, automation, and
sustainability. A structured literature review was conducted using the PRISMA framework to review peer-reviewed stud-
ies on AI applications in architectural design, urban planning, and smart cities published between 2003 and 2025. The
reviewed studies demonstrate that AI enhances generative design, streamlines spatial organization, and supports sustain-
able architecture. However, challenges such as algorithmic bias, ethical concerns, and loss of architectural identity persist.
Future research should emphasize ethical implementation, interdisciplinary collaboration, and integration with VR/AR to
enable immersive, real-time, and informed design processes.

Keywords Artificial intelligence (AI) · Augmented reality · Architectural design · Technologies

Abbreviations DRL Deep Reinforcement Learning

3DBP 3D Building Printing DSS Decision-Making Support Tools
ACO Ant Colony Optimization ES Knowledge-Based Systems
AF Additive Fabrication FNN Feedforward Neural Network
AI Artificial Intelligence NNs Feedforward Neural Networks
AL Automated Learning GAN Generative Adversarial Networks
AR Augmented Reality GD Generative Draft
BEM Building Energy Modeling GIS Geographic Information System
BIM Building Information Modeling GWO Grey Wolf Optimizer
CAD Computer-Aided Design HVAC Heating, Ventilation, and Air Conditioning
CFD Computational Fluid Dynamics (Innovative)
CNA Convolutional Neural Architectures ML Machine Learning
CNN Convolutional Neural Networks PRISMA Preferred Reporting Items for Systematic
DBMS Database Management Frameworks Reviews and Meta-Analyses
DI Digital Twin PSS Planning Assistance Platforms
DL Deep Learning R-CNN Region-Based CNN
RL Reinforcement Learning
SDG Sustainable Development Goals
SVM Support Vector Machines
Ibraheem Nooruddeen Albukhari
[email protected] VDC Virtual Design and Construction
VR Virtual Reality
1
Umm Al Qura University College of Engineering and
Architecture, Makkah, Saudi Arabia

13
 Journal of Umm Al-Qura University for Engineering and Architecture

1 Introduction [12]. Furthermore, Alnaser et al. (2024) and Ilvitskaya et

al. (2018) explore how these environments reduce ecologi-
Artificial Intelligence (AI) is increasingly redefining archi- cal footprints by enhancing energy conservation, minimiz-
tectural design by introducing data-driven methods that ing material waste, and utilizing sustainable resources while
enhance efficiency, precision, and spatial adaptability [1]. improving occupant health and comfort [9, 12].
Generative design is among its most transformative appli- Despite global efforts, challenges persist due to the unde-
cations, where algorithms autonomously generate and rutilization of new technologies, financial constraints, and
evaluate thousands of design alternatives based on defined inconsistent regulations that governments face. Many stud-
constraints such as form, function, materiality, and energy ies support the UN Sustainable Development Goals (SDGs)
performance [2]. These tools offer a departure from intu- concerning clean energy, sustainable cities, and climate
ition-based practices, allowing architects to balance struc- action [13, 14]. This research underscores the potential of
tural integrity, sustainability, and spatial innovation through emerging technologies, including AI and ML, in enhancing
computational precision [1–4]. energy efficiency. When used together, the simulation tools
Machine learning (ML) and neural networks have trans- Design Builder and EnergyPlus help architecture, engineer-
formed architectural planning by replacing intuition-based ing, and research professionals achieve optimal building
decisions with data-driven insights, optimizing space, pre- designs while meeting energy efficiency and sustainability
dicting performance, and enhancing energy efficiency [5, 6]. goals [15].
The increasing adoption of ML across various architectural Data-driven strategies and computational techniques are
applications highlights its potential to improve design accu- transforming the construction sector by enhancing produc-
racy, sustainability, and structural integrity [5, 6]. tivity through automated planning, digital infrastructure,
Despite its potential, the integration of computational simulations, and structural optimization [16].
intelligence in architectural planning remains constrained. Despite global efforts, challenges persist due to the unde-
Automated learning (AL) has been applied to design space rutilization of new technologies, financial constraints, and
exploration, blueprint generation, style adaptation, and envi- inconsistent regulations that governments face. Many stud-
ronmental evaluation, but data availability challenges hinder ies support the UN Sustainable Development Goals (SDGs)
its broader implementation [7]. While the AL approach can concerning clean energy, sustainable cities, and climate
enhance the assessment of ecological efficiency for archi- action [13, 14]. This research underscores the potential of
tects, it has been proposed that AL could be a powerful tool emerging technologies, including AI and ML, in enhancing
in significantly lowering the carbon footprint of structures. energy efficiency. When used together, the simulation tools
However, incorporating AL in architectural workflows is Design Builder and EnergyPlus help architecture, engineer-
restricted by a fundamental challenge—data availability. ing, and research professionals achieve optimal building
These computational models necessitate vast volumes of designs while meeting energy efficiency and sustainability
annotated input data to identify underlying trends within goals [15].
datasets [5, 8]. For example, an extensive collection of pre- In parallel, the global construction industry faces ongo-
viously computed cases is essential to develop an algorithm ing challenges including declining productivity, workforce
capable of accurately forecasting the lifecycle carbon emis- shortages, and rising project complexity. In response, AI
sions of a planned structure. Ideally, this dataset should be and automation have been proposed as solutions within
sourced from measurements to ensure maximum accuracy. the framework of the Fourth Industrial Revolution. These
Nevertheless, in many scenarios, obtaining such data is not include AI-driven structural optimization, automated design
feasible. Regarding carbon footprint evaluation, it is pos- generation, and robotic construction systems, each aiming
sible to digitally model various architectural forms, analyze to streamline workflows, improve resource allocation, and
their embodied and operational carbon, and utilize these address skill gaps across the sector [16]. Furthermore, AI’s
instances to train the AL system [5, 8, 9]. influence transcends architecture, as evidenced by land-
Several studies have reviewed the current applications mark achievements in other fields. For instance, AlphaGo’s
and future directions for improving energy efficiency, disruption of traditional game theory reflects the broader
emphasizing its effectiveness in homes and its role in foster- capacity of AI to recalibrate human decision-making and
ing more intelligent, sustainable environments while reduc- strategic planning. In architecture, similar paradigm shifts
ing carbon emissions [10, 11]. From a broader perspective, are unfolding as AI transitions from a supportive tool to an
a study by Alnaser et al. (2024) investigated the concept of active agent in creative and technical processes [16, 17].
a green-built environment, which incorporates eco-friendly In light of these developments, this systematic review
and resource-efficient strategies throughout a structure’s seeks to consolidate emerging research on AI in architec-
entire lifespan, from conceptualization to deconstruction tural design. By critically examining the technologies,

13
Journal of Umm Al-Qura University for Engineering and Architecture

applications, challenges, and future directions, this study 1.3 Main aim
aims to provide a coherent synthesis that informs both aca-
demic inquiry and professional practice in the era of com- This study aims to systematically examine and synthe-
putational design. size the emerging applications of artificial intelligence in
architectural design, with particular attention to how such
1.1 Research problem technologies influence creativity, sustainability, and profes-
sional collaboration across disciplinary boundaries.
In recent years, architectural design has entered a transfor-
mative phase marked by the integration of artificial intel- 1.4 Research objectives
ligence (AI) across planning, modeling, and construction
workflows. Although multiple technologies have been intro- This study seeks to comprehensively examine the role of
duced and theoretical contributions continue to expand, a AI in contemporary architectural design by systematically
fragmented understanding persists regarding the systematic reviewing prevailing technologies and their practical imple-
role of AI in addressing contemporary architectural chal- mentations. The primary objectives are as follows:
lenges such as energy performance, spatial optimization,
and creative automation. Disciplinary fragmentation, soft- ● To identify and categorize the principal AI methodolo-
ware incompatibility, and educational deficiencies further gies, including machine learning, deep learning, neural
complicate the development of a unified AI design frame- networks, and generative design, currently employed
work. As a result, there remains an urgent need to consoli- within architectural design and planning processes.
date technological advances, practical applications, and ● To investigate how AI augments fundamental architec-
interdisciplinary insights to evaluate how AI can be effec- tural design practices, such as spatial configuration, ma-
tively operationalized within the built environment. terial optimization, structural analysis, energy efficiency
modeling, and parametric design strategies.
1.2 Research questions ● To assess the convergence of AI with emerging digital
technologies, notably virtual reality (VR), augmented
To achieve these objectives, the following research ques- reality (AR), 3D printing, and building information mod-
tions are addressed: eling (BIM), and its influence on architectural visualiza-
tion, design automation, and construction workflows.
● What are the most significant AI technologies currently ● To critically evaluate the technical constraints and im-
utilized in architectural design, and how do they contrib- plementation challenges associated with the adoption
ute to the field’s evolution? of AI in architectural contexts, including issues related
● In what ways does AI enhance creativity and facili- to data availability, algorithmic complexity, and system
tate more effective decision-making in architectural integration.
processes? ● To examine the technical, pedagogical, and ethi-
● What are the primary challenges associated with AI- cal challenges limiting the broader adoption of AI in
driven architecture, and how can these obstacles be ad- architecture.
dressed to maximize AI’s potential in the industry? ● To propose future research directions that respond to
● How is AI expected to influence the future of architec- current knowledge gaps and promote inclusive, collab-
tural design, and what emerging trends will shape its de- orative frameworks for AI integration in architecture.
velopment in the coming years?
1.5 The evolution of AI in architectural design
How is artificial intelligence currently applied in archi-
tectural design, and in what ways do these technologies Since ancient times, people have tried to create systems
enhance generative creativity, environmental performance, that mimic how the human brain works. The implemen-
and interdisciplinary integration within contemporary tation of AI has evolved logically over time through con-
design practice? By addressing these questions, this review tinuous advancements in mathematical reasoning, physical
aims to provide a comprehensive overview of how AI is sciences, and philosophical thought, alongside rapid prog-
reshaping architectural design and offer insights into this ress in computational technology [18]. Computational
technology’s potential future directions. intelligence encompasses a broad field that examines and
enhances human thinking through its fundamental principles
and technology-driven components. AI systems have been
continuously developing for more than sixty years before

13
 Journal of Umm Al-Qura University for Engineering and Architecture

achieving widespread application across various domains, methods remained relatively outdated before the rapid
with architectural planning being one of the primary areas advancement of AI. Studies focused solely on theoretical AI
of implementation. Computational intelligence started as models without practical applications were excluded, as this
a design tool in architecture but has become a partner for review emphasizes real-world implementations. Research
architects. It could lead to the decision-making process [19]. unrelated to architecture—such as work in mechanical engi-
Architecture has a long history of computational inno- neering or robotics—was also omitted. Additionally, papers
vation, from the earliest computer-aided design (CAD) lacking sufficient experimental validation were excluded to
software to the current use of parametric design tools. ensure credibility. These criteria narrowed the review to two
Nevertheless, the actual revolutionization of architectural high-quality studies demonstrating robust evidence of AI-
workflows came due to the advent of ML and deep learning driven architectural processes.
(DL) technologies [20]. Unlike rule-based systems, AI algo-
rithms can learn from vast datasets and generate new design 2.2 Study selection
solutions by identifying patterns and trends, making them
dynamic and versatile. A particular form of AI, generative This review adheres to the Preferred Reporting Items for
adversarial networks (GANs), has been widely popular in Systematic Reviews and Meta-Analyses (PRISMA) frame-
architecture because they can generate new design concepts work to ensure transparency and reproducibility in study
with human-like creativity [21]. The shift to AI-powered selection. The PRISMA flow diagram (Fig. 1) delineates
methodologies for architectural design is quite profound, the stepwise filtration process, beginning with the identi-
and this means moving from static computational tools to fication of 1,500 records across databases such as Web of
them. For instance, using historical data, ML is utilized Science, Scopus, and Google Scholar. Following title and
in predictive modeling with AI systems predicting future abstract screening, 450 full-text articles were evaluated
design needs or problems in a particular field. Additionally, against defined inclusion criteria. Studies were selected
AI can automatically optimize building layouts and configu- based on their empirical application of AI in architectural
rations to maximize space utilization and energy consump- design, publication between 2003 and 2024, and relevance
tion, providing more terrific design refinement that is not to generative, sustainable, or urban-scale design. Exclusion
feasible with traditional methods [22]. criteria included theoretical-only papers, non-English lan-
guage publications, and research not specific to architecture.
Data extraction followed a structured protocol encompass-
2 Research methodology ing AI methods employed, performance metrics, application
domains, and outcome relevance. Quality assessment was
2.1 Inclusion and exclusion criteria conducted using journal ranking indicators, peer-review sta-
tus, and evidence of experimental validation, ensuring that
For this review, the studies were selected based on a rig- only high-impact, methodologically sound studies informed
orous set of inclusion and exclusion criteria to ensure the synthesis.
relevance and quality. The search terms included “AI in
architecture,” “generative design in architecture,” “ML in
urban planning,” “neural networks for sustainable architec- 3 Literature review
ture,” “AI-driven smart cities,” and “automation in architec-
tural processes.” 3.1 Core AI technologies in architectural practice
The inclusion criteria prioritized peer-reviewed jour-
nal articles, conference proceedings, and case studies that Machine Learning (ML) encompasses three main catego-
offered substantial contributions to understanding AI in ries: supervised, unsupervised, and reinforcement learning.
architectural design. The selected studies were published In supervised learning, algorithms perform classification
between 2003 and 2025, a period marked by significant (e.g., decision trees, support vector machines (SVMs) and
advancements in AI-driven methodologies. The review regression (e.g., linear regression, neural networks), recog-
focused on research applying AI to architectural design, nizing patterns and making informed predictions based on
including Building Information Modeling (BIM) and sus- labeled datasets [23]. Unsupervised learning, by contrast,
tainable construction. Empirical studies and real-world case identifies patterns in unlabeled data using clustering tech-
studies were included to provide concrete evidence of AI’s niques such as centroid-based, mean-shift, and probabilistic
practical impact. clustering models. Reinforcement learning optimizes deci-
On the contrary, studies that did not align with the objec- sion-making through trial-and-error approaches [23].
tives of this review were excluded. Computational design

13
Journal of Umm Al-Qura University for Engineering and Architecture

Fig. 1 PRISMA Flow Diagram

These ML techniques have been widely implemented cases, ideally based on empirical measurements. In many
in intelligent energy management systems, significantly scenarios, such data is not readily available. In some cases,
enhancing energy efficiency in buildings [24]. Research- digital modeling of architectural forms and the analysis of
ers have demonstrated that ML algorithms optimize energy their embodied and operational carbon can create the data
consumption, contributing to improved sustainability and needed to train AL systems [8, 9].
cost-effectiveness in architectural projects. Automated According to various studies [26, 27] comparing differ-
learning (AL), a branch of computational intelligence, ent assessment and certification systems, the categories that
relies on statistical principles rather than explicit program- consistently carry the highest weights include environmen-
ming. It identifies recurring trends within extensive datasets tal loading, internal environment, external environmental
and develops mathematical frameworks representing data impact, and resource use. Generally, environmental factors
dynamics. Different categories of AL exist; some rely on are considered more relevant than socioeconomic and polit-
predefined training datasets, while others acquire knowl- ical aspects, as well as planning and innovation [26, 27].
edge independently, without human intervention. Neural The weightings of these systems are determined by the
architectures, particularly Convolutional Neural Architec- degree of importance assigned to each category of evalua-
tures (CNA) [8], use convolutional layers to enhance visual tion, and these priorities are directly shaped by the specific
data analysis, processing images through neural networks contexts and challenges of each country. Among the vari-
with input, output, and hidden layers [8, 25]. ous systems, energy efficiency requirements consistently
Despite its potential, AL integration in architectural plan- receive higher weighting, reflecting its recognition as one
ning is constrained by data availability challenges. AL has of the most urgent needs in sustainable architectural design.
been applied to design space exploration, blueprint gen- Other high-priority categories include material sustainabil-
eration, style adaptation, and environmental evaluation [7]. ity, potable water conservation, and parameters that influ-
However, these computational models require vast volumes ence indoor environmental comfort, such as visual, thermal,
of annotated data to identify underlying trends accurately and acoustic comfort, as well as indoor air quality [26–28].
[8]. For instance, predicting lifecycle carbon emissions Based on these weightings and taking into account
requires an extensive collection of previously computed the architects’ thought process during design, five key

13
 Journal of Umm Al-Qura University for Engineering and Architecture

sustainability indicators were identified: energy efficiency, on leveraging ML optimization to predict the compressive
potable water conservation, indoor air quality, acous- strength of eco-friendly concrete or geopolymer concrete,
tic comfort, and project durability. These indicators were an environmentally sustainable alternative to traditional
deemed the most representative and suitable for integration cement-based concrete. The research utilizes a feedforward
into the methodological framework of architectural design neural network (FNN) enhanced by the bird-inspired opti-
practices [29]. mization technique, the Grey Wolf Optimizer (GWO), to
Numerous studies highlight the contributions of AI to refine predictive accuracy [36]. These studies underscore
architectural optimization. Chew et al. (2024) applied deep AI’s multifaceted role in advancing architectural planning,
learning (DL) techniques to predict structural integrity, sig- construction, and heritage conservation.
nificantly reducing design iteration time [30]. Sharma et al. A comprehensive summary of the key AI technologies,
(2024) explored the use of ML in automated zoning anal- their applications, challenges, and future prospects in archi-
ysis, demonstrating its potential to enhance the efficiency tectural design and planning is presented in Table 1 below.
of architectural design [31]. AI has also proven valuable
in space utilization, rapidly generating new spatial layouts 3.2 Generative design and computational creativity
based on thousands of existing layouts [32]. These methods in architecture design and planning
can theoretically solve spatial utilization problems faster
than manual design. Nonetheless, challenges such as inter- Generative design (GD) and computational creativity are
pretability and data bias persist, necessitating human over- increasingly transforming architectural practice by leverag-
sight in AI-driven design [32]. ing data-driven strategies to improve productivity through
Building performance predictions increasingly rely on automated planning, digital infrastructure, simulations, and
predictive modeling that incorporates climatic, structural, structural optimization [16]. Recent research has focused on
and user behavioral data [33]. AI-driven methods for space generative modeling, which minimizes manual labor and
optimization enhance efficiency and improve spatial layout time by automatically generating diverse design alterna-
materiality, while adaptive building structures enable archi- tives. This approach streamlines repetitive tasks, enhances
tects to design dynamic frameworks that respond to envi- operational efficiency, and supports industry professionals
ronmental changes [34]. These advancements collectively by using algorithmic frameworks to deliver practical, smart,
contribute to the evolution of intelligent and responsive and customized solutions [37].
architectural designs [34]. The reviewed literature includes studies examining
Rodrigues et al. [35] discuss the incorporation of DL with Autodesk Generative Design’s impact on parametric archi-
3D building modeling and Building Information Modeling tecture [38]. According to a study [39], AI has already been
(BIM) to automate structural degradation classification in used in adaptive facades that minimize energy consump-
historic buildings. They use a region with CNN features tion through real-time environmental data analysis. Despite
(R-CNN) that has been trained using images that outline these advancements, AI-generated designs often lack intui-
architectural anomalies. At the same time, a laser scanning tive human creativity, necessitating architectural refinement
process creates a point cloud reference to support the Build- and human interpretation before implementation [39].
ing Information Modeling (BIM) model. This new approach GD has rapidly gained traction in architectural research,
automates the detection and classification of structural faults, reflecting its growing role in academic and professional
relating the geometric and numerical information with BIM practice [27]. GD involves creating multiple design options,
to support facility management [35]. The workflow has four evaluating their performance, and refining them through
significant steps: data acquisition through terrestrial laser optimization. Marsh (2008) focused on GD’s performance,
scanning, modeling in 3D using BIM, image processing for while other researchers emphasized the process of generat-
DL-based defect identification, and incorporation of these ing, evaluating, and iteratively improving design options.
results into Revit software. The research highlights the abil- These studies often classify GD according to either perfor-
ity of digital technologies and AI to update conservation mance outcomes or process methodologies. In this context,
activities and deliver a structured decision-support system GD development is typically analyzed based on algorithms
for architectural heritage upkeep. The primary challenge and design constraints [29]. Algorithms act as programming
met is the automatic detection of anomalies, previously con- instructions for creating and refining design alternatives.
ducted manually, that ensures more trustworthy and trans- Common approaches include evolutionary techniques,
parent documentation within a digital ecosystem based on shape-based design rules, fractal patterns, swarm intel-
BIM [35]. ligence, and cellular growth models. The selection of an
Furthermore, ML techniques have been applied to sus- algorithm depends on the project goal, with shape-based
tainable building materials. Wang et al. [36] present a study rules suited for space layout exploration and evolutionary

13
Journal of Umm Al-Qura University for Engineering and Architecture

Table 1 AI technologies and its applications in architecture design and planning

AI Key application & Advantages Challenges Functionality Real-world Future prospects Ref.
technology applications
Machine Predictive modeling enhances It requires exten- Analyzes large Used in smart Expected to [24,
Learning energy efficiency, reduces costs, and sive, high-quality datasets to identify energy grids to improve automa- 30–
(ML) improves urban planning through datasets, but patterns and make optimize HVAC tion in sustainable 32,
energy consumption forecasting, there is potential data-driven deci- systems and construction and 34,
structural safety assessments, auto- data bias. sions for architec- predict material intelligent building 36]
mated zoning analysis, and sustain- tural design and lifespan. management.
able material optimization. urban planning.
Neural Net- Analysis of smart city infrastruc- High computa- Processes complex, Applied in carbon Future develop- [8,
works (NNs) ture, lifecycle carbon emissions tional demand high-dimensional footprint analysis ments may lead to 9,
forecasting, and predictive mod- requires exten- data to identify cor- of buildings and real-time carbon 12]
eling for building performance. sive labeled data. relations between climate-respon- tracking in urban
Enhances automation and data-driven architectural factors. sive architectural environments.
decision-making. designs.
Deep Learn- Predicting structural integrity, clas- Computation- Uses multi-layered Employed in Expected to [30,
ing (DL) sifying historic building defects, ally expensive neural networks to automated fault enhance predictive 31,
and integrating BIM improve design and requires analyze archi- detection in maintenance and 35,
efficiency and structural analysis significant train- tectural projects’ heritage building AI-driven architec- 36]
accuracy. ing data. images, structures, conservation. tural restoration.
and materials.
Automated Design space exploration, environ- It requires Learns from past Utilized in smart Future AL systems [8,
Learning mental evaluation, energy efficiency vast annotated design iterations city projects to may autonomously 9]
(AL) analysis, and blueprint genera- datasets, but data and optimizes improve urban generate green
tion. Reduces carbon footprint and availability has energy-efficient infrastructure building plans.
enhances ecological efficiency. limitations. architecture. design.
Convolu- Visual data analysis, defect detection, Requires large Extracts features Used in drone- Integration with [8,
tional Neural and classification in historic build- image datasets from architectural based struc- AR for real-time 9,
Networks ings. High accuracy in pattern recog- and computa- images for pattern tural damage defect detection. 35]
(CNNs) nition and predictive maintenance. tional power. identification. assessment.
Reinforce- Adaptive zoning regulations, land-use Computation- Uses trial-and-error Implemented Future AI-driven
ment Learn- optimization, and AI-driven space ally intensive; learning to optimize smart zoning land-use models [31–
ing (RL) optimization. Improves efficiency requires complex land-use planning. tools for urban may predict and 33]
in land allocation and regulatory simulation development. prevent overpopu-
compliance. environments. lation stress.
Generative AI AI-driven space layout optimization It may limit Generates new Applied in AI- It could evolve into [27,
based on existing designs. Faster human creativ- design variations generated urban AI-powered collab- 29,
design iterations and optimized ity and require based on learned layouts for hous- orative design tools 37,
layouts. extensive train- architectural styles. ing developments. for architects. 38]
ing data.
Feedfor- Predicting compressive strength of Requires high- Processes input data Used in optimiz- Future improve-
ward Neural eco-friendly concrete (geopolymer quality experi- in layers to predict ing sustainable ments could lead [36]
Networks concrete). Enhances material sustain- mental data for material properties. construction to AI-driven eco-
(FNNs) ability and construction efficiency. training. materials. friendly material
selection.
Region- Structural defect classification and High computa- Detects and clas- Employed in AI- Expected advance-
Based CNN automation of conservation activi- tional require- sifies architectural powered heritage ments may allow [35]
(R-CNN) ties in historic buildings. Improves ments and a need defects using DL. site restoration. real-time AI-driven
accuracy in anomaly detection and for extensive conservation
documentation. labeled image efforts.
datasets.
Optimization Enhancing predictive accuracy in sus- Requires careful Uses nature-inspired Applied in mate- Future iterations
Algorithms tainable material strength forecasting. parameter tuning metaheuristic tech- rial property may enhance [36]
(Grey Wolf Optimizes material performance and and validation. niques to optimize forecasting for predictive model-
Optimizer reduces experimental costs. predictions. eco-friendly ing in innovative
- GWO) buildings. materials.

techniques better suited for design optimization [40]. tools are being developed to translate textual constraints
Design constraints establish boundaries by defining param- into digital formats, most research still relies on manually
eters such as dimensions, materials, construction methods, inputting these constraints for computational processing.
cost, and spatial considerations. While some automation Further studies are needed to explore GD development and

13
 Journal of Umm Al-Qura University for Engineering and Architecture

its components to improve understanding of their relation- 3.3 AI in sustainable architecture and energy
ships. GD is a step-by-step process in which designers input optimization
their goals, algorithms, and constraints into the system,
generating continuously adjusted models until an optimal AI has emerged as a critical enabler in advancing sustain-
design is achieved [27, 40]. able architectural design, particularly through its capacity to
Intelligent planning and construction have become key optimize energy consumption, material selection, and envi-
driving forces in the building sector. Numerous design ronmental performance. A growing body of literature high-
enhancement investigations have been carried out, utilizing lights how AI tools, when integrated into design workflows,
computational intelligence, data-driven methodologies, and support architects in simulating ecological conditions, ana-
optimization principles. This growing research has fueled lyzing historical energy data, and generating low-carbon
interest in generative modeling, which autonomously pro- material solutions [45].
duces algorithm-driven design variations, significantly min- ML models have been employed to analyze consumption
imizing human involvement and time consumption [41]. patterns in high-rise buildings, offering predictive insights
Comparable to other branches of engineering, auto- for HVAC optimization and operational efficiency. For
operation is progressively taking a crucial part in architec- instance, recent studies demonstrate the utility of AI in cali-
ture as developments in AI optimize multiple procedures brating building systems based on historical occupancy and
and improve design potential. AI-driven auto-operation is weather data, enabling more adaptive and energy-resilient
explored in various facets of architectural planning, encom- designs. Tari et al. (2024) demonstrated that AI-based com-
passing digital building representation, virtual design and putational fluid dynamics (CFD) simulations could also be
construction (VDC), intelligent 3D modeling, algorithmic applied to optimize ventilation in green buildings, reducing
design, computational design, efficiency assessment and cooling loads to similar levels [46]. Furthermore, the inte-
simulation, workflow coordination, report creation and gration of AI with the Internet of Things (IoT) enables real-
assessment, and digital manufacturing [42]. Regarding the time energy regulation through dynamic sensor feedback
latter, auto-operation revolutionizes how different archi- and automated system adjustments [47]. In addition, various
tectural structures and components are conceptualized and AI-powered tools are used to assess the carbon footprint and
produced. Sophisticated additive fabrication (AF) tech- recyclability of construction materials [48].
niques, such as three-dimensional printing (3DP), enable Urban growth has led to increased energy demand in
creators to apply swift model development approaches and buildings, contributing significantly to CO₂ emissions. The
construct intricate and tailored elements directly, minimiz- construction sector accounts for nearly 30% of global energy
ing resource wastage and building duration [43]. Further- consumption, highlighting substantial potential for effi-
more, auto-operation leveraging AI and 3DP is extensively ciency improvements [49]. The building envelope is crucial
studied across various domains utilizing diverse substances for energy savings, and Building Energy Modeling (BEM)
and fabrication scales, including metal-based 3DP methods tools aid in optimizing designs. Despite advancements in
for aviation and mechanical sectors, components produced predictive modeling, challenges such as inconsistencies
from composite materials through photopolymer-based between optimized models and architectural plans persist.
3DP, along with advancements in biomedical substances via Traditional design methods depend on iterative modifica-
AI-driven three-dimensional and four-dimensional printing tions, which limit efficiency [49]. To address these limita-
technology research [42]. tions, researchers propose hybrid strategies that integrate
AI-driven approaches, including generative adversarial ML algorithms with evolutionary optimization techniques.
networks (GANs), deep reinforcement learning (DRL), and This combination enhances the accuracy and adaptability of
ML, are transforming architecture by advancing adaptive simulations across diverse building typologies by incorpo-
facades, architectural design, structural analysis, and intel- rating climatic, structural, and user behavior variables [47,
ligent materials and robotics that enhance durability and 49–51]. These approaches show promise in generating con-
performance [44]. Collectively, these studies illustrate how text-specific energy solutions. However, several challenges
generative design and computational creativity are reshap- hinder widespread implementation. High initial costs make
ing architectural workflows, opening new avenues for it difficult for some organizations to adopt AI technologies,
innovation and sustainability (as detailed in Table 2, which while the need for advanced computational infrastructure
summarizes AI-based methods, their applications, benefits, and concerns over data privacy, particularly in systems that
and associated challenges). collect and process large volumes of sensitive information,
remain significant barriers [52].
Ultimately, while AI offers substantial potential to
transform sustainable architectural practices, its effective

13
Journal of Umm Al-Qura University for Engineering and Architecture

Table 2 AI-Based methods for generative design in architecture: applications, benefits, and challenges
AI-based method Use case Description Key contributions Benefits Challenges Ref.
Generative Adaptive Uses DL models to gener- Automates facade Energy-efficient High computation
Adversarial Net- Facades ate architectural designs designs for energy- structures, reduction in cost, complex training [44]
works (GANs) that adapt to environmental efficient structures. energy costs, dynamic process
conditions. environmental response
Deep Reinforce- Urban AI-driven simulations for Reduces manual Faster zoning approv- Complexity in execu-
ment Learning Planning optimizing land use, zon- effort in zoning als, optimized land tion, need for extensive [45,
(DRL) Automation ing, and infrastructure. and enhances use, enhanced urban training data 46]
sustainability. sustainability
Parametric AI Automated Uses AI-driven rules to Improves efficiency Customizable layouts, Limited aesthetic intu-
Tools Blueprint generate architectural blue- in large-scale reduction in material ition, high dependency [27,
Generation prints and floor plans. planning and waste, improved work- on input parameters 29]
customization. flow efficiency
AI-Driven Struc- Earthquake- Simulates structural integ- Enhances predic- Enhanced safety Dependence on exten-
tural Simulations Resistant rity under different condi- tive analysis for standards, predictive sive structural data, [30,
Designs tions for risk assessment. disaster-resistant maintenance, data-driven computational intensity 31]
structures. risk assessment
Smart Material Self-Adapting AI optimizes material It helps create Sustainable construc- Costly implementa-
AI Structures properties for responsive buildings that adapt tion, real-time material tion, limited material [45,
and sustainable designs. to environmental adaptation, longevity database 49]
changes. improvement
Data-Driven AI-Enhanced Uses ML algorithms to Reduces opera- Reduced material waste, Requires extensive
Optimization Building optimize energy consump- tional costs lower lifecycle costs, training data, interoper- [23,
Performance tion, material efficiency, and improves enhanced operational ability challenges 24,
and maintenance efficiency. efficiency 33]
schedules.
Machine Automated AI-based models predict Minimizes project Improved accuracy in Understanding AI’s role
Learning (ML) Construction and optimize construction delays and cost project execution, reduc- in complex decision- [31,
Algorithms Planning schedules and budgets. overruns. tion in planning errors, making requires robust 32]
cost savings training datasets.
Artificial Neural Struc- AI predicts load distribu- Improves safety Improved safety, Requires extensive
Networks tural Load tion for optimized architec- and material optimized materials, labeled datasets, [24,
(ANNs) Predictions tural designs. efficiency in enhanced performance computational power 34]
construction. assessment constraints, model
interpretability
Genetic Automated It uses evolutionary algo- Enhances space Space-efficient layouts, High computational
Algorithms Space rithms to generate optimal efficiency in adaptive designs, cost, long optimization [27,
Planning spatial layouts. urban and interior reduced planning effort time, and the require- 40]
designs. ment of expert tuning
Ant Colony Structural AI mimics nature to create Optimizes resource Nature-inspired opti- Complexity in imple-
Optimization Adaptation self-adaptive architectural utilization and mization, improved resil- mentation and lack [41,
(ACO) designs. improves resilience. ience, efficient material of standardization 42]
usage require interdisciplinary
expertise
Convolutional Architectural AI applies artistic or his- Automates design Automated design aes- Requires large labeled
Neural Networks Style Transfer torical styles to architec- aesthetics, enabling thetics, improved visual datasets, computation- [35,
(CNNs) tural renderings. quick concept realism in architectural ally expensive 36]
visualization. renderings
Evolutionary Design AI iterates thousands Reduces manual Automated iteration Long processing
Algorithms Optimization of design variations to effort and improves of design variations, time, computational [29,
find the best-performing sustainability. energy-efficient solutions complexity 37]
solution.
Reinforcement Construction AI trains robots to auto- Enhances automa- Improved automation Data-heavy training
Learning (RL) Robotics mate construction tasks tion in complex in on-site construction, process, risk of unpre- [55,
and improve efficiency. construction tasks. real-time adaptability dictable outcomes 56]
AI-Driven Digital Build- AI optimizes multiple Enhances automa- Improved design effi- High computa-
Auto-Operation ing Represen- architectural planning tion in architec- ciency, reduced waste, tional cost, need for [57,
tation, VDC, processes, including tural workflows faster modeling, and domain expertise, 58]
3D Modeling workflow coordination, and structural production and challenges in
computational design, and optimization. interoperability
efficiency assessment.

13
 Journal of Umm Al-Qura University for Engineering and Architecture

application depends on addressing infrastructural, regula- showed DL models that can accurately predict congestion
tory, and methodological constraints. The reviewed lit- points and suggest the best road network designs. Similarly,
erature underscores the necessity of continued research to the authors proposed the application of reinforcement learn-
refine these technologies and support their integration into ing to dynamic traffic light control that can potentially result
mainstream architectural workflows [52]. in a 23% reduction of average travel time in a simulated
Table 3 below illustrates various AI-driven applications urban environment [4]. Quan (2022) proposed the applica-
in sustainable architecture and energy optimization, out- tion of reinforcement learning and an autoencoder for the
lining the technologies employed, associated benefits, and traffic light control task in an urban environment, reducing
implementation challenges. travel time by 12.72% for the optimal control strategy [55].
Over time, city development has integrated various
3.4 AI in urban design and smart city planning sophisticated analytical methods, with GIS being the most
commonly utilized. Other innovations, such as database
Urban AI refers to integrating AI into city environments, management frameworks (DBMS), decision-making sup-
building on past technological advancements, and aligning port tools (DSS), planning assistance platforms (PSS), and
with the smart city agenda. AI-driven systems are trans- knowledge-based systems (ES), have experienced different
forming urban life through applications such as intelligent degrees of recognition [56]. While AI and intelligent urban
assistants, autonomous vehicles, AI-powered travel, and technologies have reshaped sectors like commerce and sup-
surveillance. These technologies redefine how people inter- ply chain management, city planning has been slower in
act with spaces and raise ethical concerns about privacy, adopting these advancements.
democracy, and governance [53]. As technology has advanced, flexible building skins
With its capacity for data-driven decision-making, AI is have become an interdisciplinary field of study, integrat-
revolutionizing urban planning by enabling smart city infra- ing computational and AI methods to enhance adaptability.
structure development. Advanced tools such as big data Automated systems and advanced sensors optimize ventila-
analytics, Geographic Information System (GIS) mapping, tion and shading, reducing energy consumption. AI-driven
and AI-based predictive models help city planners optimize process integration further streamlines the design workflow,
land use, transportation networks, and environmental sus- promoting environmentally responsible and sustainable
tainability [54] (see Table 4). A study by del Campo & Man- architectural solutions [57].
ninger (2024) on using AI in real-time traffic monitoring

Table 3 AI-Driven sustain- AI application Use case Technology Benefits Challenges Ref.
able architecture and energy used
optimization
AI-Powered Climate-responsive building ML, Reduction in High compu-
Energy designs, including optimizing predictive energy costs tational cost [23,
Simulations orientation, glazing, and shading modeling 24]
strategies for energy efficiency
Innovative Adaptive temperature regulation IoT sen- Enhanced Requires
Heating, Venti- based on real-time data, seasonal sors, rein- occupant extensive [47,
lation, and Air adjustments, and occupancy forcement comfort training data 49]
Conditioning patterns to optimize comfort and learning
(HVAC) Control energy use
Systems
AI-Based Smart glass and shading systems Neural Reduced light- Cost of
Daylighting that adjust based on natural light networks, ing energy sensor [48]
Optimization availability to reduce reliance on sensor demand integration
artificial lighting fusion
Neural Dynamic insulation systems that DL, Lower heat- Requires
Network-Based adapt based on climate conditions thermal ing/cooling large-scale [46,
Insulation and building usage, optimizing modeling expenses adoption 49]
Models thermal performance
AI-Driven Car- Green material selection, optimiza- AI-driven Reduced Limited
bon Footprint tion of supply chain to minimize lifecycle environmental availability [12,
Analysis emissions, and lifecycle analysis analysis, impact of databases 14]
for long-term sustainability big data
Automated AI-based recycling of materials, Computer Sustainable Complex
Waste Reduc- waste sorting, and optimized waste vision, construction logistics [50,
tion Algorithms management strategies for sustain- robotics and waste 52]
able construction automation segregation

13
Journal of Umm Al-Qura University for Engineering and Architecture

Table 4 AI applications in urban planning and smart cities: methods, benefits, and challenges
AI method Application Advantages Challenges Ref.
DL for Traffic Smart mobility Reduced congestion, improved traffic flow, better High computational requirements, real-time [54,
Prediction systems route planning through predictive analytics data processing constraints, and data privacy 55]
concerns
Reinforcement Adaptive traffic Optimized signal timing, reduced vehicle idle time, Data collection complexity, potential overfit- [55,
Learning (RL) for light control lower emissions, dynamic learning adaptation ting in different traffic conditions, high 56]
Traffic Signals computational cost
AI-Based Land Zoning and Efficient space utilization, improved urban design, Algorithmic bias concerns, lack of trans- [31,
Use Planning infrastructure better infrastructure planning, reduced urban parency in decision-making, difficulty in 57]
optimization sprawl. integrating diverse data sources
Neural Networks Air quality Real-time pollution tracking, better environmental Sensor network dependency, calibration, and [47,
for Pollution prediction and policies, early warning systems for public health maintenance costs, potential data inaccura- 49]
Monitoring tracking protection cies due to environmental changes
AI-Driven Disas- Flood, earth- Increased disaster preparedness, faster emergency Data reliability issues, integration with early [58,
ter Risk Analysis quake, and response, reduced casualties, improved risk warning systems, computational cost of 59]
climate disaster mitigation large-scale simulations
prediction
Smart Grid AI-assisted Sustainable power management, reduced energy Integration challenges with existing grid [47,
Optimization energy waste, enhanced grid resilience, predictive infrastructure, cybersecurity risks in AI- 48]
distribution maintenance powered grids, regulatory concerns
Computer AI-driven Automated anomaly detection, enhanced public Privacy concerns, ethical issues related to [60,
Vision for Smart security and safety, reduced emergency response time, real-time surveillance, risk of bias in AI detection 61]
Surveillance monitoring crime detection models
AI-Powered Waste Intelligent Optimized waste disposal routes, reduced landfill Sensor accuracy issues, high implemen- [50,
Management waste collection overflow, improved recycling rates through AI tation costs, adoption barriers in urban 52]
and recycling sorting infrastructure
optimization
Big Data Analytics Urban infra- Enhances decision-making with real-time insights, Data security concerns, integration with [57,
structure improves urban services efficiency, facilitates legacy systems, challenges in processing 58]
planning predictive analytics for city development and storing massive datasets
Geographic Infor- Land use and Spatial analysis for better urban planning, envi- Requires constant data updates, high cost of [12,
mation Systems transporta- ronmental sustainability, and improved disaster implementation, challenges in integrating 59]
(GIS) AI tion network preparedness with real-time AI models
optimization
Database Manage- Urban data stor- Efficiently manages and processes large datasets, Security vulnerabilities, potential data [59,
ment Systems age and retrieval supports real-time decision-making, enables AI- silos, need for robust data governance 60]
(DBMS) & AI driven automation in city management frameworks.
Decision Support Wise city policy It aids policymakers with data-driven insights, It requires high-quality data, has potential [57,
Systems (DSS) and planning supports urban sustainability goals, and improves bias in AI models, and has difficulty inte- 58]
& AI long-term urban management strategies. grating qualitative and quantitative data.
Planning Assis- Urban develop- Provides strategic insights for city growth, Needs integration with other AI tools, chal- [57,
tance Platforms ment projects enhances stakeholder collaboration, and supports lenges in stakeholder adoption, and com- 58]
(PSS) AI scenario-based urban modeling plexity in urban governance frameworks
Knowledge-Based Smart city Supports automated decision-making, improves Knowledge acquisition is complex; there is [56–
Systems (ES) governance regulatory compliance, and enhances city adminis- a risk of outdated decision rules and chal- 58]
& AI tration efficiency lenges in AI explainability.

According to various research papers, AI can also be construction while preserving structural integrity and aes-
incorporated into disaster-resilient architecture to predict thetic quality [60].
earthquake resistance, flood vulnerability, and wind resil- Table 4 below illustrates the diverse applications of AI
ience. Architects can use ML algorithms to proactively in urban planning and smart cities. It highlights the specific
design structures that mitigate risks, thereby supporting AI methods, their practical use cases, advantages, and the
more effective post-disaster recovery through analysis of associated implementation challenges.
historical data [58, 59].
Another emerging field is AI-driven material science,
which develops bio-based, low-carbon, and recyclable
building materials using AI recommendation systems.
These innovations help reduce the environmental impact of

13
 Journal of Umm Al-Qura University for Engineering and Architecture

4 Integration of real-world case studies in grounded and contextually adaptive AI applications in the
AI-driven architectural practice built environment [59].

Although AI technologies have been comprehensively docu- 5.1 Challenges and ethical considerations
mented in terms of technical capabilities, their practical util-
ity is most convincingly demonstrated through real-world In the rapidly evolving landscape of architectural design,
case studies. These projects serve not only as proof-of- the integration of AI presents both transformative opportu-
concept applications but also highlight the potential and nities and multifaceted ethical challenges. As design work-
limitations of AI integration within architectural workflows flows become increasingly data-driven and algorithmically
and urban contexts. Cases such as The Edge in Amsterdam enhanced, critical questions emerge regarding data sover-
illustrate the application of AI-based environmental control eignty, decision-making authority, and the future role of
systems that respond in real time to user behavior and indoor human creativity. To ensure responsible implementation,
conditions [61]. Similarly, Google’s Bay View Campus in it is essential to address concerns, including data privacy
California incorporates predictive simulation to enhance issues, security risks, algorithmic bias, and the evolving role
solar optimization and energy efficiency [62]. Hyundai of human creativity in the design process [22, 66].
Motorstudio Goyang in South Korea has adopted genera-
tive AI algorithms to refine spatial articulation and struc- 5.1.1 Data privacy, security, and interdisciplinary gaps
tural logic [63]. In heritage conservation, the Basilica of
Santa Maria di Collemaggio demonstrates how DL meth- AI-driven architecture, predictive analytics, and intelligent
ods can classify and interpret structural damage in post-seis- surveillance enhance urban planning but raise privacy and
mic contexts [64]. Notably, Al Bahr Towers in Abu Dhabi security concerns. Biometric identification and facial recog-
integrates AI into a responsive façade system that combines nition improve traffic control and security, enabling wide-
traditional aesthetics with environmental regulation. Taken spread surveillance. Ethical dilemmas arise as personal data
together, these cases reflect an emerging shift toward com- often lacks clear standards for collection, storage, and use
putational methods that not only enhance performance out- [66].
comes but also challenge conventional boundaries between AI-assisted building automation enhances smart homes
form, function, and sustainability [65]. and offices but poses security risks by collecting private
user data. Continuous data collection for lighting, climate,
and security systems increases vulnerability to cyberattacks,
5 Discussion potentially allowing unauthorized access to homes and busi-
nesses [56].
The integration of AI architectural design signifies a funda- Consequently, AI-driven architecture should incorporate
mental shift in design methodology [1]. As evidenced in the the principles of privacy by design to mitigate the risks,
literature, AI is increasingly regarded not as a supplemen- which involves transparent, secure, and regulated data col-
tary tool but as a central mechanism for enabling generative lection. To balance innovation with ethical responsibility,
creativity, performance-driven optimization, and sustain- robust cybersecurity protocols, encryption methods, user
able innovation. Techniques such as ML, DL, and GANs consent mechanisms, and other suitable measures must be
support predictive modeling and algorithmic workflows, prioritized [56].
allowing for more adaptive and data-informed design pro- The principle of “privacy by design” necessitates that
cesses [1, 2]. data protection considerations be embedded into the archi-
The following discussion critically engages with the tectural design process from inception rather than as retro-
principal challenges and ethical considerations surround- active safeguards. For instance, smart buildings equipped
ing AI integration in architectural design, including con- with AI-driven occupancy sensors and behavioral monitor-
cerns related to data privacy, algorithmic bias, and the ing systems must incorporate anonymization protocols, data
evolving role of human creativity [59]. It further distin- minimization strategies, and user consent mechanisms at
guishes between AI methodologies and conventional com- the software and hardware levels [67]. A practical applica-
putational approaches, underscoring their conceptual and tion can be seen in intelligent lighting systems that adjust
functional divergences. Moreover, the discussion outlines according to movement but store only anonymized meta-
future research trajectories and strategic recommendations, data, thereby eliminating the risk of personal identifica-
emphasizing the importance of climate-responsive design, tion. Additionally, facial recognition technologies, when
immersive technologies such as VR/AR, heritage preserva- deployed in urban security or access control systems, must
tion, and interdisciplinary collaboration to ensure ethically operate under transparent governance models that clearly

13
Journal of Umm Al-Qura University for Engineering and Architecture

define data retention periods, access rights, and regulatory software, which can generate elaborate and optimized struc-
oversight [67]. tures that would otherwise take human architects months to
The effective integration of AI in architectural practice develop. While AI does the heavy lifting of these designs,
is often impeded by enduring interdisciplinary gaps. While these designs often lack any sense of aesthetic subjectivity
architects are primarily educated in spatial reasoning, aes- or contextual understanding, and it is uncertain if AI can
thetic judgment, and user-centered design, they frequently replace human creativity or if it only ups the quality of it [71].
lack proficiency in algorithmic logic, data structures, and Moreover, over-reliance on AI in architecture risks reducing
computational modeling, which are core competencies architects to mere supervisors of algorithmic outputs rather
essential to AI development. This asymmetry in domain than active creators. The human layer, though, may not be
expertise presents a significant barrier to collaborative inno- involved in the ideation of these very plausible yet untested
vation and underscores the need for cross-disciplinary lit- solutions, which could have disastrous outcomes—they did
eracy and integrated design-computation pedagogies [68]. not account for the historical, cultural, and environmental
layers that should exist in place of the AI [71].
5.1.2 Algorithmic bias and decision-making in AI The fear that AI may diminish the role of architects to
architecture the point of rendering the profession obsolete and folding
it into computer science is unfounded and reductive. Archi-
A significant concern in architectural design is the potential tecture is not solely a problem-solving enterprise driven by
bias in AI models. Historical datasets are the training data- data or algorithms; it is a profoundly humanistic and con-
sets AI uses to learn, and if the datasets are skewed or not textual practice. Architects operate at the intersection of
representative, they will reinforce preexisting inequalities in design, society, culture, and the environment. While AI can
design and accessibility [3, 4]. A critical inquiry emerging in assist in the production and evaluation of designs, it cannot
the context of computational intelligence is whether it pos- assume authorship in the way human designers can, as it
sesses the capacity to replicate or potentially surpass human lacks the consciousness, intent, and experiential knowledge
creativity within the architectural discipline. AI demon- necessary for creating spaces with identity, empathy, and
strates powerful capabilities in optimization, generative meaning [69].
modeling, and structural simulation. However, despite these Rather than dissolving the profession, AI challenges
advancements, it remains inherently limited in its capac- architects to redefine their roles as creative leaders, ethical
ity to interpret cultural contexts, ethical dilemmas, and the stewards, and interdisciplinary collaborators. Far from dis-
emotional and symbolic nuances that are integral to archi- placing architects, AI requires them to engage more deeply
tectural authorship [3]. with technology, with human needs, and with the broader
systems in which architecture operates. This evolution will
5.1.3 Human creativity vs. AI-generated design not relegate architecture to a branch of computer science but
will instead enrich it through synergy [69].
Research indicates that architectural decision-making is AI should be embraced not as a replacement for the
not purely rational, as architects often settle for satisfactory human mind but as a generative partner that extends the
solutions rather than seeking the optimal design. Various boundaries of design thinking. Through thoughtful integra-
human and social factors influence this tendency. Cognitive tion, architects can harness the computational power of AI
biases frequently shape architectural choices, prompting the while maintaining authority over conceptual vision, cultural
introduction of structured methods to evaluate their impact. expression, and societal relevance. This future affirms the
Studies have examined numerous cognitive biases concern- enduring significance of architectural education and profes-
ing the contextual factors that influence architectural deci- sional practice, reshaped, not erased, by the intelligent tools
sions [69]. we develop [69].
AI-driven architecture is one of the most debated topics,
particularly regarding the role of human creativity versus 5.1.4 Differences between AI and traditional
AI-generated design. AI shines in data-driven and opti- computational design
mization-driven design, generative design, and structural
analysis, but falls short in the ways that the most innately AI and other traditional computational methods are often
human, cultural, intuitive, and emotional architects push a confused as the same, but many applications and qualities
building to exist beyond the results of technical problem- differ. Conventional computational design has supported
solving [70]. architects by generating and modifying designs using pre-
Visualization of complex, highly optimized structures can defined rules, parametric modeling, and deterministic algo-
be created by AI-powered tools such as parametric design rithms. The methods listed here are strongly programmed

13
 Journal of Umm Al-Qura University for Engineering and Architecture

logic, in which the outcome is defined within the human comparative overview of AI-driven and traditional architec-
definition and not with an independent learning or decision- tural design methods, highlighting key distinctions in data
making process [72]. integration, design strategies, and sustainability. AI-Driven
On the other hand, AI-powered design systems utilize Design scores very high (5) in areas like Data Analysis,
ML, neural networks, and generating algorithms to explore Automation & Speed, Integration with Tech, and Emerging-
voracious datasets, recognize patterns, and develop the best Tech, showing strong performance in modern, tech-focused
feasible solutions that go contrary to the fixed rules. In con- domains. It scores moderately (3–4) in Creativity, Ethics,
trast to conventional computational tools in which the user and Heritage. In contrast, Traditional Methods score higher
explicitly specifies the qualification of every design modi- in Heritage & Restoration (4) and Ethical Considerations
fication, AI models are trained on data, design according (4), but remain moderate to low (2–3) in most other cat-
to contextual constraints, and autonomously create novel egories, especially in tech-related aspects. This highlights
architectural solutions [1, 2]. The most superficial way of AI’s strength in innovation and efficiency, while traditional
expressing this is in the use of generic and reinforcement methods retain value in human and cultural elements.
learning algorithms to optimize architectural models based
on AI, whose formulations define the knowledge encoded in 5.2 Future research directions
these models and allow them to refine themselves, provid-
ing solutions that even the most creative human designer As AI advances, its role in architectural design will expand
may not use due to complexity or computational handicap far beyond its current applications, leading to innovative
[1, 2]. Parametric design and energy efficiency optimization ways to conserve energy, reduce emissions, and enhance the
are sufficient evidence of AI’s superiority over conventional creative process. Future research should focus on the fol-
computational design. AI-driven software simulates models, lowing key areas to maximize AI’s impact while addressing
predicts environmental impact, analyzes user behavior, and ethical and practical challenges.
dynamically adapts designs to improve sustainability and
functionality [1]. AI-based systems can also be adaptable 5.2.1 Digital transformation and industry innovation
in real time to change structures based on live sensor data,
a feature that static computational models cannot do [1, 2]. The Smart Industry Revolution is reshaping architec-
To illustrate these differences more clearly, Fig. 2 presents a ture and design in the construction industry through rapid

Fig. 2 A radar chart comparing AI-Driven Design and Traditional Methods across ten design criteria, using a 1–5 scale

13
Journal of Umm Al-Qura University for Engineering and Architecture

technological advancements and the increasing availability optimizing energy efficiency, and promoting sustainable
of electronic data. The adoption of digital tools has surged, materials, it can also contribute to advancing climate-
mainly due to the COVID-19 pandemic, which accelerated responsive design (Table 5). Future research should focus
mobile and cloud computing use. While some companies on AI-driven parametric design tools that generate struc-
have struggled, those embracing digital technologies, such tures based on real-time climate data. These tools would
as BIM and Digital Twin (DT), are poised to emerge stron- allow architects to develop self-adjusting facades, passive
ger. These technologies help improve efficiency, sustain- cooling systems, and solar-optimized layouts with little or
ability, and collaboration across the design and construction no energy consumption [58].
phases. As the industry shifts towards these innovations, With the intensifying effects of climate change and the
smaller firms face challenges in adopting such technolo- global call for decarbonized construction, climate-respon-
gies due to cost barriers. Nonetheless, digital solutions are sive and disaster-resilient architecture is no longer a design
becoming vital for ensuring safety, monitoring progress, ambition but a necessary standard. AI systems capable
and facilitating construction work [15, 35, 73]. The future of simulating environmental variables in real time offer
of architecture and construction design is centered on these promising avenues for optimizing building performance
transformative technologies, which are expected to drive in response to dynamic climatic conditions [58]. Research
growth, enhance project management, and improve over- is required to explore how reinforcement learning, when
all outcomes. Future research will focus on identifying applied to structural adaptation models, can predict and
disruptive and emerging technologies, analyzing lifecycle respond to extreme weather events, energy fluctuations, or
management, and exploring sustainability in architectural material stress. These technologies must be integrated with
design and construction [73]. parametric design tools that allow architectural elements to
adapt autonomously. However, such tools will require an
5.2.2 Urban planning and digital twins ecosystem of high-fidelity environmental data, responsive
feedback loops, and domain-specific calibration strategies,
Urban planning, as another frontier of AI application, is areas that remain underdeveloped [58, 59].
transitioning toward real-time responsiveness. While smart
cities have incorporated elements of AI, such as traffic 5.2.4 AI, VR, and AR in immersive design
prediction and energy load balancing, more comprehen-
sive models, particularly those using digital twins, are still With AI, VR, and Augmented Reality (AR) integration,
emerging. Future studies must investigate the role of AI in architectural visualization and design processes can poten-
predictive zoning, multi-objective land-use planning, and tially have a new lease of life. VR and AR technologies
infrastructure forecasting [15, 35]. The urban twin concept enable architects, clients, and stakeholders to experience
should evolve from static replication to dynamic prediction and interact with existing and novel designs in immersive
models that simulate behavioral, environmental, and infra- environments augmented with real-time data and video
structural changes. This can support policymakers in evalu- feeds. The following research step will determine how AI
ating urban interventions before implementation. However, can improve real-time simulating models within VR and
success will depend on the availability of interoperable AR platforms so architects can experiment with design
urban datasets and cross-sector collaboration between data variations in real time [74]. These could be generative AI-
scientists, municipal authorities, and architectural designers powered models that change designs per user preferences,
[73]. environment considerations, and structural constraints. In
addition, the existing progress in haptic feedback and fore-
5.2.3 Climate-responsive and disaster-resilient architecture sighted AI exploration could somehow boost the practicality
of architectural walkthroughs through VR. There is another
All AI research should emphasize climate-responsive and opportunity to explore using AI-enabled AR for on-site con-
disaster-resilient architecture to support environmentally struction assistance. AR interfaces, augmented by AI, could
responsible urban development [59]. add digital models over construction, helping architects and
Architecture integrates innovative solutions like adap- engineers see through the physical design, catch inconsis-
tive building façades to address climate change and sus- tencies, ensure spatial accuracy, and cut material waste [74].
tainability. These intelligent systems regulate temperature, In parallel, the accelerated development of Extended
ventilation, and solar radiation, enhancing energy efficiency Reality (XR), encompassing Virtual Reality (VR), Aug-
and comfort through innovative materials and automation. mented Reality (AR), and Mixed Reality (MR), has intro-
Just as AI plays a central role in various climate-respon- duced new modalities of architectural visualization. Yet, the
sive design practices, such as assessing weather patterns, research community has not fully addressed the potential

13
 Journal of Umm Al-Qura University for Engineering and Architecture

Table 5 Future directions of AI in Research Key applications Potential Challenges & Future research directions Ref.
architecture and construction focus benefits Considerations
AI, VR, - AI-powered - Improved - High computa- - Advancing AI-driven
and AR in generative design design visu- tional costs generative models for [62,
Architecture in VR/AR alization and - Integration adaptive design 74]
- Real-time design iteration with existing - Enhancing real-time AR
simulations - Reduced workflows overlays for construction
- AI-assisted construction - Ethical validation
spatial planning errors concerns in - AI-driven automated
- AR-guided - Enhanced AI-generated code compliance verifica-
construction collaboration aesthetics tion for VR-based designs
workflows and client
interaction
AI-Assisted - AI-driven mate- - Preserving - Ensuring - Developing AI models
Renovation rial analysis for cultural heritage AI-generated trained on diverse histori- [59,
& Heritage aging structures - Reducing reconstructions cal architecture datasets 64]
Conservation - Automated dam- restoration costs remain histori- - Integrating AI with
age assessment - Predicting cally accurate 3D printing for precise
via AI-powered long-term struc- - Ethical con- restoration
drones tural integrity cerns in altering - AI-driven predictive
- AI-assisted res- heritage sites algorithms for climate
toration planning impact on historical sites
AI’s Role - AI-driven para- - Enhanced - Limited avail- - Advancing AI tools
in Climate- metric climate energy ability of high- for climate-adaptive [45,
Responsive modeling efficiency quality climate architecture 49]
& Resilient - Self-regulating - Reduced car- data - AI-assisted selection
Architecture building facades bon footprint - Balancing sus- of bio-based, recyclable
- AI-powered - Better disaster tainability with materials
disaster resilience preparedness cost-effectiveness - AI-integrated urban heat
simulations and mitigation island effect mitigation
strategies
AI for - AI-optimized - Reduced - Managing - Integrating AI with
Smart Cities zoning and land urban large-scale urban IoT for real-time urban [54,
and Urban use analysis congestion data sets monitoring 55]
Planning - AI-powered traf- - More efficient - Ethical - AI-assisted green city
fic flow modeling use of resources concerns in planning with renewable
- Predictive main- - Improved pub- AI-driven energy
tenance for urban lic safety and decision-making - AI-driven simulation
infrastructure sustainability models for optimizing
urban resilience against
extreme weather
AI in - AI-optimized - Reduced con- - Limited - AI-based material
Sustainable selection of low- struction waste adoption of AI science for biodegrad- [49,
Construction carbon materials - Increased in traditional able and self-repairing 52]
Materials - AI-driven durability of construction materials
material lifecycle materials - Need for regu- - AI-powered predictive
analysis - Cost-effective latory approval analysis for long-term
- AI-assisted 3D and sustain- of AI-designed durability
printing of eco- able building materials - AI-driven structural
friendly structures practices optimization for reducing
embodied carbon

of merging XR platforms with generative AI systems [74]. interactive design experience — a concept that remains
Future studies must examine the extent to which AI can largely theoretical but is urgently worth investigating [74].
support real-time spatial manipulation in immersive envi-
ronments. In this context, generative algorithms could act 5.2.5 Heritage conservation and AI-assisted renovation
as co-design agents, interpreting verbal, visual, or sensory
inputs from users to iteratively refine design proposals. In the context of heritage conservation, AI presents oppor-
When further complemented by haptic feedback and AI- tunities for automated damage assessment, predictive aging
informed material simulations, the result could be a fully models, and data-informed restoration planning—yet fur-
ther inquiry is needed to ensure historical fidelity and

13
Journal of Umm Al-Qura University for Engineering and Architecture

ethical intervention [35]. Climate-resilient architecture also ethical reasoning, and systems literacy to prepare students
remains underexplored, particularly concerning AI-driven not just to use AI, but to critique and shape it. Likewise,
adaptive façades, passive system optimization, and real-time the research community must focus on the development of
risk modeling for natural disasters. Additionally, develop- regulatory protocols that ensure transparency, inclusive-
ing open-access datasets and interoperable platforms would ness, and accountability in AI-generated design decisions.
enhance the practical viability of AI tools in architectural Without such frameworks, the integration of AI into archi-
practice. Addressing these gaps requires cross-disciplinary tecture risks being driven by market dynamics rather than
frameworks that bring together designers, data scientists, public benefit [75–78].
historians, and environmental engineers to collaboratively
shape a technologically advanced yet culturally responsive 5.3 Limitations
built environment [59].
Structural deterioration, material limitations, and the Despite adhering to a structured and rigorous review meth-
necessity of maintaining authenticity in restoring these odology, this study is subject to certain limitations that war-
buildings and cultural heritage sites present particular pres- rant consideration. The exclusion of non-English literature
ervation challenges. In such cases, AI can revolutionize may have inadvertently limited the scope of the review by
heritage conservation by analyzing architectural styles, pre- omitting regionally significant developments and culturally
dicting material aging, and creating restoration plans based specific implementations of AI in architecture. Furthermore,
on the first architectural style with modern safety standards the reliance on established academic databases such as Web
while retaining structural integrity [35]. of Science, Scopus, and Google Scholar, while ensuring the
Table 5 below outlines the future research directions, inclusion of high-impact peer-reviewed sources, may have
benefits, applications, and challenges associated with AI resulted in the omission of valuable insights published in
integration across architecture and construction subfields, niche journals, conference proceedings, or industry-specific
including heritage conservation, climate-responsive design, repositories not indexed in these platforms. As such, the
smart cities, and sustainable materials. synthesized findings, although reflective of prevailing aca-
demic discourse, may not fully encapsulate the diversity of
5.2.6 Intelligent materials and predictive modeling global practices or emerging perspectives situated outside
mainstream publication channels.
From a materials perspective, the emergence of bio-based
and recyclable composites has broadened the possibili-
ties for sustainable construction. However, these materi- 6 Conclusion and recommendations
als often lack robust data on long-term performance under
varied stress conditions. AI has the capacity to bridge this This research highlights a promising avenue for future
gap through predictive modeling [1]. Future research should work: developing AI algorithms trained on historical archi-
address the integration of AI in evaluating lifecycle per- tectural data. These models can help reconstruct missing
formance, toxicity levels, and thermal efficiency of next- design elements, recommend sustainable restoration materi-
generation construction materials. Optimization algorithms als, and simulate environmental impacts on heritage build-
such as genetic programming and swarm intelligence could ings. AI-powered drones and robotic scanning systems can
assist in identifying the ideal composition of hybrid mate- produce high-resolution 3D maps of endangered sites, sup-
rials [1, 2]. This, however, demands a multidisciplinary porting conservationists in damage assessment and preser-
research model where material scientists, architects, and vation planning.
computer engineers operate within a unified computational However, such research must address the ethical con-
framework. siderations of using AI to recreate heritage. While AI can
improve restoration accuracy, interventions must be guided
5.2.7 Governance, education, and ethical frameworks by cultural integrity and historical authenticity rather than
modern biases or stylistic assumptions.
Lastly, governance, education, and public trust will play a AI in architecture enhances efficiency, creativity, and
decisive role in AI’s architectural future. The professional sustainability by optimizing workflows and structures. To
divide between architects and technologists can no longer maximize these benefits and mitigate risks, AI models must
be sustained. It must be acknowledged that the introduc- be transparent, unbiased, and inclusive. Policymakers and
tion of AI into design practice necessitates a recalibration industry leaders should establish ethical guidelines to pre-
of pedagogical content across architectural institutions [75]. vent architectural homogeneity, exclusion, and privacy con-
Future curricula should incorporate algorithmic thinking, cerns. AI-driven designs must respect diverse cultural and

13
 Journal of Umm Al-Qura University for Engineering and Architecture

2. Nagy D et al (2017) Project discover: An application of genera-

social needs, upholding architectural integrity and ensuring tive design for architectural space planning. In Proceedings of the
inclusivity. Symposium on Simulation for Architecture and Urban Design
Importantly, AI should not be seen as a replacement for 3. Kudless A (2023) Hierarchies of bias in artificial intelligence
human architects, but rather as a creative partner that aug- architecture: collective, computational, and cognitive. Int J Archi-
tectural Comput 21(2):256–279
ments their work. Incorporating AI training into architec- 4. del Campo M, Manninger S, AI AND THE MORPHOLOGY
tural education will help future professionals understand OF THE CITY AI’S, DYNAMIC POWER IN SHAPING THE
how to use AI tools effectively while respecting artistic and (2024). The Routledge Companion to Smart Design Thinking in
cultural values. Architecture & Urbanism for a Sustainable, Living Planet,: p. 86
5. Płoszaj-Mazurek M, Ryńska E, Grochulska-Salak M (2020)
All AI research should emphasize climate-responsive Methods to optimize carbon footprint of buildings in regenera-
and disaster-resilient architecture to promote environmen- tive architectural design with the use of machine learning, con-
tally responsible urban development. AI-based solutions volutional neural network, and parametric design. Energies
can improve energy efficiency, optimize material usage, and 13(20):5289
6. Valls F et al (2018) Urban data and urban design: A data min-
support the design of sustainable cities, ultimately reducing ing approach to architecture education. Telematics Inform
the environmental impact of architectural projects. 35(4):1039–1052
Future work should also focus on integrating AI with 7. Rumbarger J, Vitullo R (2003) Architectural graphic standards
emerging technologies such as VR, AR, and robotics. These for residential construction: the architect’s and builder’s guide to
design, planning, and construction details, vol 4. Wiley
combined technologies can create more immersive, efficient, 8. Abreu S (2019) Automated architecture design for deep neural
and innovative design processes, transforming architectural networks. ArXiv Preprint ArXiv:190810714
design into a dynamic canvas for creativity. Finally, strong 9. Ilvitskaya S, Prihodko V (2018) Innovative technologies in the
security and privacy frameworks are essential for AI-driven field of topography, land management, territorial planning, con-
struction and architecture. in IOP Conference Series: Materials
smart buildings. The future of AI in architecture depends on Science and Engineering. IOP Publishing
its ethical integration, enhancing design, and enriching the 10. Alyami M (2022) An approach to enhancing energy performance
built environment. in residential buildings in hot climate regions: the case of Saudi
Arabia. University of Nottingham
Author contributions This review was written solely by Dr. Ibraheem. 11. Asif M (2016) Growth and sustainability trends in the buildings
sector in the GCC region with particular reference to the KSA and
UAE. Renew Sustain Energy Rev 55:1267–1273
Funding The review did not receive any funding.
12. Alnaser AA, Maxi M, Elmousalami H (2024) AI-Powered digital
twins and internet of things for smart cities and sustainable Build-
Data availability Data will be made available upon request. ing environment. Appl Sci 14(24):12056
13. Kutty AA et al (2020) A system thinking approach for harmo-
Declarations nizing smart and sustainable City initiatives with united nations
sustainable development goals. Sustain Dev 28(5):1347–1365
14. Okeke FO (2021) Analysing the role of architects in achieving the
Conflict of interest The authors declare no conflicts of interest.
sustainable cities goal by 2030. University-Led Knowledge and
Innovation for Sustainable Development. Edited by F. A Anyogu,
Open Access This article is licensed under a Creative Commons CA Eme, J. A Ogbodo published by Boldscholar Research Ltd.
Attribution 4.0 International License, which permits use, sharing, Abuja, Nigeria.:104– 12
adaptation, distribution and reproduction in any medium or format, 15. Xu X, Mumford T, Zou PX (2021) Life-cycle Building informa-
as long as you give appropriate credit to the original author(s) and the tion modelling (BIM) engaged framework for improving Build-
source, provide a link to the Creative Commons licence, and indicate ing energy performance. Energy Build 231:110496
if changes were made. The images or other third party material in this 16. Lee J et al (2023) Simplified methods for generative design that
article are included in the article’s Creative Commons licence, unless combine evaluation techniques for automated conceptual Build-
indicated otherwise in a credit line to the material. If material is not ing design. Appl Sci 13(23):12856
included in the article’s Creative Commons licence and your intended 17. Bölek B, Tutal O, Özbaşaran H (2023) A systematic review on
use is not permitted by statutory regulation or exceeds the permitted artificial intelligence applications in architecture. J Des Resil
use, you will need to obtain permission directly from the copyright Archit Plann 4(1):91–104
holder. To view a copy of this licence, visit ​h​t​t​p​​:​/​/​​c​r​e​a​​t​i​​v​e​c​​o​m​m​o​​n​s​.​​o​ 18. Meeran A, AI and, Architecture (2021) Image-Based machine
r​g​​/​l​i​c​e​n​s​e​s​/​b​y​/​4​.​0​/. learning for Early-Stage design conceptualisation. Singapore
University of Technology and Design Singapore
19. Borglund C (2022) Artificial Intelligence in Architecture and its
References Impact on Design Creativity: A Study on how Artificial Intelli-
gence Affect Creativity in the Design Process
20. Hunde BR, Woldeyohannes AD (2022) Future prospects of com-
1. Kallayil A, Patadiya J, Kandasubramanian B, Adamtsevich A, puter-aided design (CAD)–A review from the perspective of arti-
Kchaou M, Aldawood FK (2025) Adaptive Smart Materials in ficial intelligence (AI), extended reality, and 3D printing. Results
Architecture: Enhancing Durability and Sustainability in Modern Eng 14:100478
Construction. ACS Omega 21. Newton D (2019) Generative deep learning in architectural
design. Technology| Archit + Des 3(2):176–189

13
Journal of Umm Al-Qura University for Engineering and Architecture

22. Allam Z, Dhunny ZA (2019) On big data, artificial intelligence application in the Building industry. Int J Constructed Environ
and smart cities. Cities 89:80–91 9(1):57–81
23. Morales EF, Escalante HJ (2022) A brief introduction to super- 44. Malik S, Tyagi AK, Mahajan S (2022) Architecture, generative
vised, unsupervised, and reinforcement learning. Biosignal model, and deep reinforcement learning for IoT applications:
processing and classification using computational learning and deep learning perspective. Artif Intelligence-based Internet
intelligence. Elsevier, pp 111–129 Things Syst.:243–265
24. Ali DMTE, Motuzienė V, Džiugaitė-Tumėnienė R (2024) Ai- 45. Krausková V, Pifko H (2021) Use of artificial intelligence in
driven innovations in Building energy management systems: A the field of sustainable architecture: current knowledge. ALFA
review of potential applications and energy savings. Energies 26:20–29
17(17):4277 46. Tari Z, Li K, Wu H (2024) Algorithms and Architectures for Par-
25. Khan A et al (2020) A survey of the recent architectures of deep allel Processing: 23rd International Conference, ICA3PP 2023,
convolutional neural networks. Artif Intell Rev 53:5455–5516 Tianjin, China, October 20–22, 2023, Proceedings, Part III. Vol.
26. Shooshtarian S (2015) Socio-economic factors for the percep- 14489.: Springer Nature
tion of outdoor thermal environments: towards Climate-sensitive 47. Shafique MA, Cross-Layer (2024) Approach to Energy-Efficient
urban design. Global Built Environ Rev.;9(3) and Secure EdgeAI: Architectures, Systems and Applications.
27. Monizza GP, Bendetti C, Matt DT (2018) Parametric and genera- in 2024 5th CPSSI International Symposium on Cyber-Physical
tive design techniques in mass-production environments as effec- Systems (Applications and Theory)(CPSAT). IEEE
tive enablers of industry 4.0 approaches in the Building industry. 48. Rane N, Choudhary S, Rane J (2023) Integrating ChatGPT, Bard,
Autom Constr 92:270–285 and leading-edge generative artificial intelligence in architectural
28. Sheth D (2017) Water efficient technologies for green buildings. design and engineering: Applications, framework, and chal-
Int J Eng Innov Sci Res 1(3):5–10 lenges. Framework, and Challenges27, (November 2023
29. Marsh A (2008) Generative and performative design: a challeng- 49. Long LD (2023) An AI-driven model for predicting and opti-
ing new role for modern architects. Design mizing energy-efficient Building envelopes. Alexandria Eng J
30. Chew ZX et al (2024) Generative design in the built environment. 79:480–501
Autom Constr 166:105638 50. Etim B et al (2024) Machine learning-based modeling for struc-
31. Sharma AS, Hota H (2024) Development of clinical decision tural engineering: A comprehensive survey and applications over-
support system using genetically optimized artificial neural net- view. Buildings 14(11):3515
work. Machine learning for real world applications. Springer, pp 51. Toussi HE (2020) The application of evolutionary, generative,
261–288 and hybrid approaches in architecture design optimization. NEU
32. Indrayani L, Sangka Y (2025) Implementation of simple additive J Fac Archit (NEU-JFA) 2(2):1–20
weighting (SAW) method for selecting a tutoring center. J Artif 52. Abioye SO et al (2021) Artificial intelligence in the construc-
Intell Eng Appl (JAIEA) 4(2):1169–1173 tion industry: A review of present status, opportunities and future
33. Ïpek E, McKee SA, Caruana R, de Supinski BR, Schulz M (2006) challenges. J Building Eng 44:103299
Efficiently exploring architectural design spaces via predictive 53. Jha AK, Ghimire A, Thapa S, Jha AM, Raj R A review of AI for
modeling. ACM SIGOPS Operating Syst Rev 40(5):195–206 urban planning: Towards building sustainable smart cities. In
34. Carta S (2022) Machine learning and the city: applications in 2021 6th International Conference on Inventive Computation
architecture and urban design. Wiley Technologies (ICICT) 2021 Jan 20 (pp. 937–944). IEEE
35. Rodrigues F et al (2022) Application of deep learning approach 54. Mohanty A, Mohapatra AG, Mohanty SK (2025) Real-time traffic
for the classification of buildings’ degradation state in a BIM monitoring with AI in smart cities. Internet of vehicles and com-
methodology. Appl Sci 12(15):7403 puter vision solutions for smart City transformations. Springer, pp
36. Wang X et al (2024) Digital industrial design method in archi- 135–165
tectural design by machine learning optimization: towards sus- 55. Quan SJ (2022) Urban-GAN: an artificial intelligence-aided com-
tainable construction practices of geopolymer concrete. Buildings putation system for plural urban design. Environ Plann B: Urban
14(12):3998 Analytics City Sci 49(9):2500–2515
37. Vermeulen D, Ayoubi ME (2020) Using generative design in con- 56. Mylrea M, Gourisetti SNG (2017) Cybersecurity and optimiza-
struction applications. CS323296 Class Handout, Autodesk tion in smart autonomous buildings. Autonomy and Artificial
38. Nasir O, Kamal MA (2023) Exploring the role of parametric Intelligence: A Threat or Savior?: pp. 263–294
architecture in building design: An inclusive approach. Facta 57. Vazquez E, Randall C, Duarte JP (2019) Shape-changing archi-
Universitatis, Series: Architecture and Civil Engineering. Jun tectural skins: A review on materials, design and fabrication strat-
22:095–114 egies and performance analysis. J Facade Des Eng 7(2):93–114
39. Li Y, Chen H, Yu P, Yang L (2025) A review of artificial intel- 58. Musa MA (2016) (pp. 12–20) Effects of architectural stability,
ligence applications in architectural design: Energy-Saving reno- architectural concept and building elements on disaster resilient
vations and adaptive Building envelopes. Energies 18(4):918 in architecture. InProceedings of the
40. Ma W et al (2021) Generative design in Building informa- 59. Lucchi E et al (2024) Climate-responsive design practices: A
tion modelling (BIM): approaches and requirements. Sensors transdisciplinary methodology for achieving sustainable devel-
21(16):5439 opment goals in cultural and natural heritage. J Clean Prod
41. Mohammadi S, Tavakolan M, Zahraie B (2022) An intelligent 457:142431
simulation-based framework for automated planning of con- 60. Mengesha G, Beyond Repair (2025) A Critical Review of Smart,
crete construction works. Eng Constr Architectural Manage Sustainable, and AI-Driven Strengthening Techniques for Aging
29(2):916–939 Civil Infrastructure. Sustainable, and AI-Driven Strengthening
42. Živković M, Žujović M, Milošević J (2023) Architectural Techniques for Aging Civil Infrastructure29, (April 2025 Apr 29
3d-printed structures created using artificial intelligence: A 61. Available Online ​h​t​t​p​​s​:​/​​/​a​r​c​​h​i​​n​s​p​​i​r​e​s​​.​c​o​​m​/​2​​0​2​4​​/​0​8​​/​1​2​/​​c​a​​s​e​-​​s​t​u​d​​
review of techniques and applications. Appl Sci 13(19):10671 y​-​s​​u​s​t​​a​i​n​​a​b​l​​e​-​f​e​​a​t​​u​r​e​​s​-​o​f​​-​t​h​​e​-​e​​d​g​e​-​i​n​-​a​m​s​t​e​r​d​a​m​/
43. Van Woensel RN, van Oirschot T, Burgmans MJ, Mohammadi 62. MAZHARI ABADI NA, KURTCA G, TUGRUL AZ (2019) San
M, Hermans K (2018) Printing architecture: an overview of exist- Francisco bay-working campus
ing and promising additive manufacturing methods and their

13
 Journal of Umm Al-Qura University for Engineering and Architecture

63. Available Online ​h​t​t​p​s​:​​​/​​/​p​a​r​a​m​​e​t​r​​i​​c​-​​a​r​c​h​​i​t​​e​c​t​​u​r​​e​​.​c​​o​​m​/​h​​y​u​n​​d​​a​i​​-​m​ 73. Berčič T, Bohanec M, Ažman L, Momirski (2024) Integrating
o​​t​o​r​​s​t​u​​​d​i​o​​-​g​o​​y​​a​​n​g​-​​2​​0​1​7​/​?​s​r​s​l​t​i​d multi-criteria decision models in smart urban planning: A case
64. Aste N, Della Torre S, Adhikari RS, Buzzetti M, Del Pero C, study of architectural and urban design competitions. Smart Cit-
Leonforte F, Manfren M (2016) Sustainable church heating: the ies 7(2):786–805
Basilica Di Collemaggio case-study. Energy Build 116:218–231 74. Arranz-Paraíso S, Arranz-Paraíso D (2024) Artificial Intelligence,
65. Attia S (2017) Evaluation of adaptive facades: the case study of al Virtual Reality, and Augmented Reality to Enhance Visual Qual-
Bahr towers in the UAE. QScience Connect 2017(2):6 ity and Accessibility in Sustainable Built Environments. InDesign
66. Adam M, Hammoudeh M, Alrawashdeh R, Alsulaimy B (2024) A Strategies for Efficient and Sustainable Building Facilities. IGI
survey on security, privacy, trust, and architectural challenges in Global, pp 179–199
IoT systems. IEEE Access. Mar 28 75. Winfield AF, Jirotka M (2018) Ethical governance is essen-
67. Chia SY (2023) Analysing the Design of Privacy-Preserving tial to Building trust in robotics and artificial intelligence sys-
Data-Sharing Architecture (Doctoral dissertation, University of tems. Philosophical Trans Royal Soc A: Math Phys Eng Sci
New South Wales (Australia)) 376(2133):20180085
68. Tuomela A The Integration of Conversational AI System into 76. Ahmad K, Maabreh M, Ghaly M, Khan K, Qadir J, Al-Fuqaha
Enterprise Architecture.2025 A (2022) Developing future human-centered smart cities: criti-
69. Zalewski A, Borowa K, Ratkowski A (2017) On cognitive biases cal analysis of smart City security, data management, and ethical
in architecture decision making. in Software Architecture: 11th challenges. Comput Sci Rev 43:100452
European Conference, ECSA 2017, Canterbury, UK, September 77. Sanchez TW, Brenman M, Ye X (2025) The ethical concerns of
11–15, 2017, Proceedings 11. Springer artificial intelligence in urban planning. J Am Plann Association
70. Wu Z et al (2021) AI creativity and the human-AI co-creation 91(2):294–307
model. in Human-computer interaction. theory, methods and 78. Riedmann-Streitz C, Streitz N, Antona M, Marcus A, Margetis G,
tools: thematic area, HCI 2021, held as part of the 23rd HCI inter- Ntoa S, Rau PL, Rosenzweig E (2025) How to create and foster
national conference, hCII 2021, virtual event, July 24–29, 2021, sustainable smart cities? Insights on ethics, trust, privacy, trans-
proceedings, part i 23. Springer parency, incentives, and success. Int J Human–Computer Interact
71. O’Toole K, Horvát E-Á (2024) Extending human creativity with 41(4):2491–2522
AI. J Creativity 34(2):100080
72. Kolarevic B (2003) Architecture in the digital age. Design and Publisher’s note Springer Nature remains neutral with regard to juris-
manufacturing. Spon Press-Taylor & Francis Group, Nueva dictional claims in published maps and institutional affiliations.
York-Londres

13