See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/378855052

Artificial Intelligence in Architecture: Integration into Architectural Design

Process

Article in Engineering Research Journal · March 2024

DOI: 10.21608/erj.2024.344313

CITATIONS READS
0 1,518

2 authors, including:

Nevine Gado
Modern Academy
8 PUBLICATIONS 1 CITATION

SEE PROFILE

All content following this page was uploaded by Nevine Gado on 08 August 2024.

The user has requested enhancement of the downloaded file.

 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

Artificial Intelligence in Architecture: Integration into Architectural Design

Process

Nermen M.Matter, Nevine G.Gado*

Modern Academy for Engineering and Technology-El Maadi, Cairo, Egypt

Abstract: Recently, Artificial Intelligence (AI) has permeated diverse sectors, revolutionizing
various aspects of human’s life by automating tasks traditionally performed by humans. This
includes manufacturing automation, certain educational methodologies, and the dynamics of
social media. However, despite these advancements, there are still domains where AI struggles to
fully emulate human capabilities, notably creativity, a unique human attribute. Architects can
leverage AI technologies to streamline their work, handling non-creative tasks efficiently and
allowing more time for innovative processes. Despite the wealth of data AI generates within
seconds, many architects and designers remain unfamiliar with integrating AI into their
processes. Through a comprehensive study of AI, including its patterns, decision-making
processes, data-related predictions, and event presentations, it becomes clear that AI can
significantly contribute to various phases of architectural design process, benefiting from its
continuous learning capabilities, self-evolution, and performance enhancement.

This research aims to formulate a framework that delineates the integration of AI and its patterns
in reshaping and enhancing the different phases of the architectural design process, focusing on
accomplishing non-creative tasks, ultimately saving valuable time and effort. This, in turn, can
support architects within the creative process, idea development, and achieving optimal results
while presenting innovative designs.

That, grounded in a theoretical framework that illuminates relevant literature related to AI, its
diverse types, and patterns. Additionally, it delves into a meticulous examination of the design
process stages within the context of AI's capabilities, boundaries, and its impact on the architect's
role. Highlighting culminates for the proposal of conceptualized AI system designed to function
as an architectural assistant. This system utilizes various AI patterns based on their specific goals
and capabilities, thereby providing comprehensive support throughout the architectural design
process.
Keywords:
Artificial Intelligence (AI)- Creativity- Architectural Design Process- AI Patterns.

A1
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

1- Artificial Intelligence (AI):

Artificial Intelligence (AI) is characterized by the capability of digital machines and computers
to execute specific tasks that mimic the cognitive functions of the human mind and intelligent
entities. These tasks encompass a range of abilities, including thinking and learning from past
experiences, emulating various mental processes. The overarching objective of artificial
intelligence is creation of systems endowed with intelligence, mirroring human-like learning and
comprehension. Such systems aspire to offer diverse services to their users, spanning education,
guidance, interaction, and beyond. (wenjun & malaeb, sep 2019)

1-1-Definitions and Concepts in the Field of Artificial Intelligence:

- Artificial Intelligence:
Artificial Intelligence (AI) is succinctly
defined as "a system's ability to correctly
interpret external data, to learn from such
data, and to use those learnings to achieve
specific goals and tasks through flexible
adaptation" (Haenlein, Kaplan, Tan , & Zhang
, 2019). Artificial intelligence emerged in the
1950s, marking its evolution as a vital branch
of data science with the overarching goal of
constructing machines capable of emulating
human thought processes. This journey ranges
from basic expressions to intricate models,
showcasing AI's capacity to progressively
think and adapt like humans.
- Machine Learning:
Machine Learning is a specialized field
within artificial intelligence that endeavors to
empower machines with the capability to
"learn" and autonomously enhance their
performance. This is accomplished through
the implementation of algorithms designed to
discern patterns and generate insights from the
data provided to them. It finds application in Fig (1) Difference between artificial intelligence, machine & deep
learning, data science
decision-making processes, future predictions, Source: https://www.datacamp.com/blog/how-to-learn-ai
and problem-solving, eliminating the necessity
for explicit programming.
Positioned as a subset of both artificial intelligence and data science, machine learning
exemplifies the evolution of intelligent systems capable of independent learning and adaptive
problem-solving. (Campesato, February 13, 2020).
- Deep Learning:
It is a subset of machine learning, representing one of the most advanced branches within the
realm of artificial intelligence. This sophisticated approach empowers machines to emulate
human learning and cognition, necessitating a complex structure that mirrors the neural networks

A2
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

found in the human brain. The primary objective of deep learning is to comprehend intricate
patterns, even when faced with noise and incomplete details.
- Data Science:
Data Science is a discipline focused on extracting valuable insights from available data.
Employing rigorous analytical methods, data science engages with both structured and
unstructured data. It serves as a vital tool for uncovering essential information from vast datasets,
often stored in repositories, by processing and analyzing the data. (Campesato, February 13,
2020).
1-2- Types of Artificial Intelligence:
Artificial intelligence is classified based on its capabilities and the functions it performs, it can be
classified into three different types based on its capabilities, as well as into four types based on
the functions it performs and its decision-making abilities ( Khan, 2021), as shown in figure (2).

That type of AI has no memory, and the task is specific. It does

not tell previous experience and works only with predefined
single logic. they do not store previous memory to implement
previous experience. Moreover, it is purely reactive and does
This type of AI is designed for specific not utilize pest experience.
tasks and is limited in its abilities. It excels
in one domain or task but lacks general
intelligence.

That type of AI system has limited memory that can be

referenced to future decisions. It is one kind of
machine learning model that especially learns from
General AI, in contrast to narrow AI, facts and events.
possesses human-like intelligence and
can perform a wide range of tasks and
understand information in a manner like
humans.
This is psychological terminology. This social intelligence
is applied to understand emotions. It predicts the
behavior of human reactions. Computers are equipped
with all types of emotions to read the human mind. It
does not exist.
This type of AI may surpass human-level
intelligence and perform tasks better
than experienced human experts. It is
characterized by the ability to learn,
communicate autonomously, and make
judgments. However, Super AI does not This is the application of common sense by the machine.
currently exist. All the predefined data is stored in the system, so
whenever it needs action, it works automatically. This
type of AI does not exist.

Fig (2) Artificial intelligence is classified based on functionalities and capabilities

Source: Khan, Hanif. (2021). Types of AI | Different Types of Artificial Intelligence Systems fossguru.com/types-of-ai-different-types-
of-artificial-intelligence-systems. 9. 50.

1-3- Applications of Artificial Intelligence:

AI applications often rely on the application of heuristics, which is essentially a ―rule of thumb‖
for solving a problem. In other words, it is a set of guidelines that often works to solve problems.
Such problems tend to be large and computationally complex, and frequently they cannot be
solved via straightforward algorithms. Additionally tend to embody a large amount of human
expertise, especially if tackled by strong AI methods. Some types of them are better solved by
using AI technique, whereas others are more suitable by traditional computer science approaches

A3
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

involving simple decision making or exact computations to produce solutions. (Campesato,

February 13, 2020) . Here we are some few examples:
 Genetic Algorithms: They are a specific methodology from the general field known as
evolutionary computation, which is a branch of AI wherein proposed solutions to a
problem adapt similarly as animal creatures adapt to their environments in real world.
Prescribed sets of rules to solve a problem and whose output is entirely predictable.
 Knowledge representation: AI systems that acquire and store knowledge to process it
and produce intelligent results also need the ability to identify and represent that
knowledge considered as an issue of representation becomes important when we consider
AI related problems. The choice of a representation is intrinsic to the nature of problem
solving and its understanding.
- Logic- based solutions: AI researchers have used a logic-based approach for
knowledge representation and problem-solving techniques. These applications
encompassed issues of understanding language and scene analysis as well as other AI
aspects.
- Semantic networks: Another graphical representation of knowledge. Much of the work
employing semantic networks has focused on representing knowledge and structure of
language.
 Expert Systems: Designed to solve complex problems by emulating the decision-making
abilities of a human expert in a specific domain. They use knowledge base and inference
engine to make decisions.
 Neural Computing Networks: it’s a subset of AI that mimic the structure and function
of the human brain, used in deep learning and pattern recognition tasks.
 Evolutionary computation: It is not solely concerned with optimization problems; it
relies upon the use of heuristics and representational paradigms. It allows machines to
interpret and understand visual information from the world, such as images and videos.
 Natural Language Processing (NLP): NLP focuses on enabling machines to understand
and interact with human language. It is used in chatbots, language translation, and voice
recognition systems.
 Reinforcement Learning AI: This application of AI learns through trial and error,
receiving feedback and rewards for its actions. It is commonly used in applications like
game playing and robotics.
These different types and applications of AI serve various purposes in architectural design and
can be applied to tasks ranging from data analysis to generative design and decision-making
processes (Zheng & Yuan, 2021).

2- The Patterns of Artificial Intelligence (AI):

Artificial Intelligence (AI) manifests through seven recurring patterns that are employed in
various combinations within its systems. These patterns are leveraged either individually for
specific use cases or in combination to achieve desired outcomes in AI and machine learning
systems. Understanding these seven patterns provides insights into how to effectively apply

A4
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

them, contributing to the realization of ultimate goals in the realm of artificial intelligence and
machine learning. As shown in figure (3)
The seven patterns are diverse and adaptable, reflecting the multifaceted nature of AI
applications. By comprehending and strategically applying these patterns, developers and
practitioners can navigate the complexities of AI systems, tailoring their approaches to specific
needs and objectives. The versatility of these patterns allows for innovation and customization,
paving the way for advancements in AI and contributing to the realization of its full potential.

Fig (3): The seven patterns of Ai

Source: https://www.forbes.com
accessed oct 15, 2023.

2-1- Hypersonalization Pattern: Treat each customer as an individual:

The Hypersonalization Pattern involves utilizing machine learning to construct an individualized
profile for each user, enabling the profile to continuously learn and evolve over time. This
pattern serves various purposes, such as presenting tailored content, suggesting related products,
and providing personalized recommendations.
The main goal of the Hypersonalization Pattern is to treat each person as a seperate and unique
individual. This approach users machine learning algorithms to analyze user behavior,
preferences, and interactions, allowing the system to understand and adapt to each user’s specific
needs and interests. (Walch , 2019)

2-2-Autonomous Systems Pattern: Reducing the need for manual labor:

The Autonomous Systems Pattern involves the integration of software, hardware, and virtual
systems with the ability to complete tasks, interact with the environment, and achieve goals with
minimal human intervention. This pattern is based on the application of machine learning
processes, allowing the system to autonomously understand its environment, predict the future
behavior of external elements, and plan how to navigate and interact with these variables.
The overall goal of the Autonomous Systems Pattern is to leverage technology and automation to
reduce reliance on manual labor and promote efficiency and autonomy in accomplishing of tasks
and goals. (Walch , 2019).

A5
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

2-3-Predictive Analytics & Decision Support:

Predictive analytics and decision support patterns use machine learning and cognitive techniques
to analyze past or current behavior, predict future outcomes, and help people make decisions
based on those patterns. The emphasis is on Unlike the autonomous systems pattern, where
machines perform tasks and decisions independently of each other, this pattern relies on human
input in decision-making, but provides important support to improve the decision-making
process. This approach, often referred to as "intelligence augmentation," aims to improve human
efficiency without replacing human participation.
The overarching objective of the Predictive Analytics & Decision Support pattern is to empower
humans with advanced insights derived from data analysis, fostering improved decision-making
across various domains while leveraging the adaptive capabilities of machine learning. (Walch ,
2019).

2-4-Conversational Pattern: Machines that can communicate as humans do:

Conversation patterns revolve around the interaction between machines and humans through
various forms of conversation and communication such as voice, text, illustrations, and images.
This pattern involves two-way interaction, where both machines communicate with humans and
vice versa. The overall goal of Conversation Patterns is to bridge the gap between humans and
machines in terms of communication. This fosters more natural and effective information
exchange, supports applications such as automated chatbots, content generation, and language
translation, and improves human-machine interaction. (Walch , 2019).

2-5-Pattern & Anomaly Detection:

Patterns and Anomaly Detection Patterns harness the power of machine learning to identify
patterns in your data and detect anomalies. ―Pattern matching" is a widely used technique in
artificial intelligence projects, and its effectiveness is increasingly recognized. Pattern and
Anomaly Detection The overall goal of patterns is to leverage machine learning capabilities to
efficiently identify patterns, differentiate between anomalies, and derive meaningful insights
from data. It has wide applications in various fields and contributes to improved decision
making, improved risk management, and detection of fraud in large datasets. (Walch , 2019).

2-6-Recognition Pattern:
The Recognition Pattern represents a significant stride in machine learning, particularly with the
integration of deep learning to enhance accuracy in tasks related to recognition or the
identification of objects. This pattern involves the use of machine learning and cognitive
methods to discern meaningful entities within unstructured content, spanning images, videos,
audio, text, or other forms of data.
The Recognition Pattern stands as a testament to the capabilities of machine learning,
particularly deep learning, in enhancing accuracy and expanding the range of applications in
diverse recognition tasks. (Walch , 2019).

2-7-Goal-Driven Systems:
Goal-driven system patterns leverage the power of reinforcement learning and other machine
learning techniques to equip systems with the ability to simulate scenarios, address recurring
problems, and provide real-time offers and auction bids. Although not as widely used as other

A6
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

patterns, it is rapidly gaining traction due to its potential to be applied to learning through
experience and trial and error.
Goal-oriented system patterns contribute to the evolving landscape of AI implementation with a
focus on experiential learning and optimal solutions. (Walch , 2019). Although the patterns
above identify different goals for using AI, each project or implementation is likely to exhibit
one or more of these patterns. In fact, most advanced AI applications combine patterns to achieve
desired results. It is important to be aware of which patterns are being used because they
determine how projects are executed and managed to achieve these goals. (Walch , 2019).

3- Artificial Intelligence and Architectural Design Process:

The integration of Artificial Intelligence (AI) into the architectural design process represents a
transformative approach, empowering architects with advanced tools and capabilities. In
architectural design, AI serves as a valuable ally, facilitating the creation of diverse design
variations and generating unique concepts by leveraging algorithms and inputs defined by
architects. This partnership with AI positions it as a supportive tool that architects can employ to
address the ever-evolving demands of their work.

3-1- Impact of Artificial Intelligence on the Role of the Architect in the Design Process:
Inarguably, advanced artificial intelligence (AI) design systems can learn design behaviors, yet
the architect's role remains vital in selecting the most suitable solution among numerous
alternatives (Cudzik & Radziszewski, 2018). While AI can generate various creative design
solutions, discerning between good and bad results poses a challenge. Evaluation involves two
stages:
 Objective evaluation: Depends on mechanical laws and self-logic based on culture and
aesthetics. Strict mechanical laws allow the AI to evaluate this.
 Subjective Experience Rating: Requires human intervention initially. Results can be added
to a database and AI evaluation tools can be created by applying AI algorithms. (Wang, Cai ,
& Tian, 2018)
It is premature to assert that AI will replace the architectural profession. Architects continue to
control the design process at every stage, choosing alternatives and posing questions AI should
answer. Designers' roles remain crucial as they are responsible for balancing various factors in
the design process (Borglund , 2022).
Soon, AI's role will be that of an assistant to architects, capable of solving specific problems
defined by the architect. Advanced AI systems cannot replace humans as architects; they can
assist in creating varied solutions but cannot make final decisions. The architectural engineer's
aspirations for creativity and control over the design environment are limitless. Hence, designers
must embrace significant differences between previous digital technologies and the new AI
methods applied in architectural engineering. (Gallo & wirz, 2020).

3-2-Limitations of Artificial Intelligence in Architectural Design:

Artificial intelligence (AI) can serve as an invaluable tool for architects due to its capacity to
analyze vast datasets (Borglund , 2022). Researchers recognize the evident constraints imposed
on AI in creative endeavors. Given that AI learns from input data, any errors in the data directly
impact the results (Wang, Cai , & Tian, 2018). Thus, while AI can assist humans in swiftly
understanding the world and predicting possibilities based on this foundation, it remains
incapable of true creativity and innovative thinking within the realm of architectural engineering.

A7
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

This is primarily because architectural engineering is an intricate practice, unlike a specific

question with a single correct answer or choice. The architectural process encompasses numerous
factors such as human elements, politics, social sciences, engineering constraints, environmental
standards, and art, all contributing to design choices(Gallo & wirz, 2020), These factors infuse
buildings with a soul, a quality that cannot be solely created by AI but relies on the practices and
visions of architectural engineers.
Therefore, it can be asserted that AI has a role in the design process, simulating various design
alternatives and processing vast amounts of data to retain and verify required parameters like
material use, energy consumption, exposure to light, noise exposure, and more (Borglund ,
2022). However, it cannot achieve the essence of buildings on its own. Consequently, it remains
crucial to consider creativity in the development of artificial intelligence applications (Haenlein
& Kaplan, 2019)

4- Architectural Design Process Stages in the Light of Artificial Intelligence:

The role of artificial intelligence (AI) in the redesign and execution of the architectural design
phase can be formalized by building AI systems. The system helps architects complete the design
process with expected efficiency while saving time, effort, and resources. It is particularly
characterized by continuous learning, which enables self-improvement and improves
performance and results.
However, it is essential that the architect leverages relevant domain expertise and actively
participates as an integral member of the team building this system.

4-1-Pre-design Stage:
This stage can be defined as the architectural programming phase, where the research and
decision-making processes shape the scope of the intended design work. It involves outlining
project requirements and goals, identifying stakeholders, and creating a roadmap for the design
process. The utilization of artificial intelligence in architectural programming aids in saving time,
obtaining more accurate data, and consequently, crafting a superior product.
 Stage 1 (Defining the Design Problem): During this stage, the problem and the primary
goal of the project are identified, followed by a comprehensive understanding of all its
aspects. This involves grasping the nature of the project, discerning client requirements,
determining the project's scale, and detailing all tasks until project completion.
 Role of Artificial Intelligence: Artificial intelligence can identify, aggregate, and categorize
various data types, discovering possible relationships between datasets and issuing
recognizable results (Borglund , 2022). By employing a combination of integrated artificial
intelligence patterns, this stage can be efficiently completed, encompassing the following:
 Hypersonalization Pattern: This pattern provides comprehensive information about the
project, similar projects, user preferences, customer needs, and available technology.
 Pattern & Anomaly Detection: This pattern searches for relevant references to the
project and identifies recurring patterns for such projects, assisting in determining
recurring advantages and problems.
 Predictive Analytics and Decision Support: This pattern assists in creating data models
for project parameters (environmental, natural, urban, economic, social, etc.). It aids in
exploring missing data and predicting future outcomes, facilitating better decision-
making and optimal selection.

A8
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

 Recognition Pattern: Utilizing natural language processing (NLP), this pattern helps
provide comprehensive information about proposed solutions to the studied problem.
 Stage 2 (Data and Information Documentation): In this stage, the study of existing
buildings, structures, and designs takes place to gather information, inspiration, and ideas
beneficial for the design process of a new project. This type of research involves examining
the work of other architects, analyzing design decisions that have been made, and
understanding the various factors that influenced those decisions to create an enlightened
new design that meets the specified needs and requirements of the proposed project. This
stage is crucial, as architects utilize it to develop a design rooted in the context of the
surrounding environment, responsive to user needs, and enlightened by the latest
technological and material advancements.
 The Role of Artificial Intelligence: Artificial intelligence contributes by analyzing vast
amounts of project-related data, employing machine learning algorithms to make design
choices based on real and reliable data (Cudzik & Radziszewski, 2018). Through the
integration of various artificial intelligence patterns, this stage can be accomplished
seamlessly. These patterns include:
 Hypersonalization Pattern: Using this pattern, architects can streamline the
architectural programming process and gain deeper insights into their clients' needs and
preferences.
 Pattern & Anomaly Detection: This pattern provides an analysis of relationships
between different architectural program elements, identifying potential conflicts or
contradictions. For example, an artificial intelligence system may be capable of
pinpointing instances where the size or location of a specific area conflicts with the
requirements of a particular function or activity. This assists designers in making better
decisions regarding building configurations, unit mixes, and other key aspects of building
design.
 Recognition Pattern: Artificial intelligence and machine learning aid in providing
comprehensive information about similar architectural projects, their elements, and the
technologies utilized. It also simplifies the process of gathering architectural precedents
by classifying and sorting all images. This makes it easier to efficiently categorize images
and leverage them later in the design process.

4-2- Design stage:

The design phase constitutes the second stage in the architectural design process. It involves the
development of conceptual plans for the building layout, form, and function. During the design
phase, expectations encompass creating design alternatives, evaluating precedents, and existing
designs, refining the design, and generating a foundational set of architectural drawings. The
information extracted from the preceding stages serves as inputs for this phase.
 Stage 3: Design Requirements (Development of Design Idea): In this stage, the design
concept is refined, and detailed design drawings are prepared to illustrate both 2D and 3D
design solutions. Additionally, space allocation, design patterns, and the selection of
construction systems and materials are specified and allocated.
 The Role of Artificial Intelligence: A set of artificial intelligence patterns can be relied
upon to accomplish this stage. This is achieved by integrating script-based artificial
intelligence programs into generative design tools, creating detailed conceptual images, and
simulating various design alternatives. (As, Pal, & Basu, 2018). This allows architects and

A9
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

designers to convey their ideas more effectively to clients, facilitating decision-making

regarding all details early in the design process. this stage can be accomplished seamlessly.
These patterns include:
 Pattern & Anomaly Detection: This pattern aids in recognizing recurring patterns from
previously input parameters in the results of the preceding stages. It utilizes this data,
transforming it into parameters to generate various types of outputs, allowing architects
to select their design outputs.
 Predictive Analytics and Decision Support: This pattern assists in simulating and
analyzing design alternatives, comparing them based on criteria such as efficiency,
functionality, and aesthetics. This makes it easier for architects to filter the best decisions
and choose promising options for further improvement.
 Recognition Pattern: This pattern provides simulation of numerous design alternatives
in a short time. Artificial intelligence can be used to create detailed 3D models of
buildings and environments, extending to integrate time, cost, environmental analysis,
and building lifecycle management. This facilitates architects in envisioning and
evaluating design options, as architects need inspiration through exploring multiple
choices—an essential element in the creative process.
 Stage4: Design Development Stage: At this stage, the review and approval of the design
plans take place, along with the preparation of executive drawings and detailed
specifications that will be used for the construction of the building or project.
 The Role of Artificial Intelligence: In this stage, two patterns of artificial intelligence are
employed. This involves creating more detailed and precise drawings of the design concept
that was developed during the previous stage, as follows:
 Goal-Driven Systems: This pattern assists in ensuring that the proposed design
achieves the required goals and standards.
 Recognition Pattern: This pattern provides the necessary data and information to create
detailed schematics of the proposed design. Subsequently, the machine learning model
can be trained on a large dataset of floor plans and corresponding features. This enables
the model to generate new floor plans based on a set of input parameters.
 Stage 5: Planning Stage (Review and Approval of Design Plans): In this stage, ensuring
that the proposed design complies with all relevant regulations and laws is carried out.
Additionally, it involves addressing the needs and requirements of the client, reviewing the
documents, and obtaining approval from the relevant authorities and stakeholders.
 The Role of Artificial Intelligence: Like the mentioned uses in the pre-design stage, text-
based artificial intelligence can assist in tasks such as scheduling, document management,
and project progress tracking (Chaillou, 2019), This is achieved by leveraging a variety of
its patterns.
 Goal-Driven Systems: This pattern aids in analyzing data from previous projects to
determine cost, timelines, and predict the requirements of future projects. This helps
ensure the accuracy and completeness of specifications, saving a significant amount of
time.
 Recognition Pattern: This pattern assists in completing design documents by
recognizing similar details in the database and utilizing them.
 Autonomous Systems Pattern: Self-operating systems help in finalizing design
documents, starting from hand-drawn sketches and data, transforming them into detailed
execution drawings and comprehensive design documents.

A10
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

 Predictive Analytics and Decision Support: Machine learning algorithms can be

utilized to provide estimates for project duration, resources, and expected budget
requirements.
The three previous stages together form a critical part of the architectural design process, where
artificial intelligence can significantly enhance efficiency, creativity, and the decision-making
process.

4-3-Construction Project Management:

The Construction Management phase is the final stage of the architectural engineering process,
involving supervision of the construction of the building or project to ensure it is built according
to the plans and design specifications. During the Construction Management phase, the architect
will regularly visit the construction site to monitor progress, ensuring it aligns with the design
plans and specifications. They review and approve the execution drawings and material samples,
addressing any problems or conflicts that may arise during construction.
In general, the Construction Management phase aims to ensure that the final project meets the
design intent and fulfills the needs and requirements of the client.
 Stage 6: Execution Stage: In this stage, everything previously planned is put into action,
and work schedules are sequenced correctly. Potential and anticipated problems are
identified and addressed before they arise.
 The Role of Artificial Intelligence: Artificial intelligence can be utilized in this stage by
analyzing data from previous projects to make predictions about the feasibility and
execution capabilities of the proposed project. It can also help identify potential risks that
may be encountered during the implementation process, such as the possibility of
construction delays or cost overruns (Rao, 2022), by leveraging a variety of its patterns.:
 Autonomous Systems Pattern: This pattern enables the monitoring of construction
errors, production issues, and on-site safety concerns. For instance, robots can routinely
capture three-dimensional images, compile site assessments, and generate quick, accurate
reports that can be automatically linked to the project plan. This saves time, money, and
effort. Moreover, self-driving construction equipment can be employed to perform
repetitive tasks such as drilling, bricklaying, concrete pouring, rebar bending, welding,
and more.
 Goal-Driven Systems: This pattern can be utilized by simulating workplace scenarios to
manage various construction components and in the human resource management of the
project. Using data recorded from previous projects helps estimate the workload,
facilitate workflow, update project sequencing, and manage tasks while keeping all
project stakeholders constantly informed about the project status.
 Predictive Analytics and Decision Support: This pattern enhances foresight into future
projects, allowing for the early detection of risks and faster responses to emerging
challenges during the implementation process. Dealing with these challenges promptly
ensures that they do not impede project completion or compromise its quality.
 Stage7: Evaluation Stage: In this stage, a post-occupancy evaluation is conducted,
gathering all data related to the end-users of the building (feedback loop).
 The Role of Artificial Intelligence: Artificial intelligence can be utilized to aggregate
extensive data on the actual usage, needs, and behaviors of end-users in the building. This
aids in developing new algorithms that can be applied in future projects, (Navarathna &
Malagi, 2018). This is accomplished using a range of integrated AI patterns:

A11
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

 Hypersonalization Pattern: This pattern determines the impression formed by users

about the project through a series of surveys directed at the end-users of the building.
 Conversational Pattern: This pattern aids in building communication programs with
users to understand their opinions about the building, its usability, and how well it meets
their needs.
 Autonomous Systems Pattern: This pattern can conduct assessments through images
and videos captured using mobile devices, drone footage, security sensor data, Building
Information Modeling (BIM), and other sources to identify deficiencies and features.
 Predictive Analytics and Decision Support: This pattern provides the aggregation and
analysis of extensive data on the actual usage, needs, and behaviors of end-users in the
building. This helps build new algorithms that can be used in future projects. Each
construction site becomes a potential source for artificial intelligence data. This provides
an opportunity for clients and professionals in the construction sector to study and benefit
from insights derived from the use of artificial intelligence and machine learning systems.
 Recognition Pattern: This pattern, through advanced analytics and artificial intelligence-
supported algorithms, helps form useful insights about the operation and performance of
a building. As a result, artificial intelligence can be used to identify problems before they
become serious and predict the necessary preventive maintenance schedules.
Fig (4) Shown the Framework for integration the role of AI and its patterns in reshaping and
practicing the phases of architectural design process.

A12
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

Architectural Design process

Client
Brief
Defining the Design Problem
Architect

Pre-design Stage
Architect
Research
Data and Information Documentation

Architect
Output Data
Machine learning& Neural networks&
Expert Systems

Architect x
Create
* Pattern & Anomaly Detection
Development of Design Idea
* Predictive Analytics and Decision Support
* Recognition Pattern

Architect
Design
Reinforcement Learning AI& Evolutionary
Design Development Stage
computation.

Design Stage
* Goal-Driven Systems
* Recognition Pattern Architect
Plan
Review and Approval of Design Plans
Client
* Goal-Driven Systems
* Recognition Pattern
* Autonomous Systems Pattern Design parameters
* Predictive Analytics and Decision Support

Architect
Output final
Deigns. + client
+experts

Architect
Execute
Construction Management stage

Client

Architect
Evaluate + client
+ experts
+ users

Project objectives

Architect
Output project + client
+ experts
+ users

Stored Data

Activity Decision Flow of Activity

Data Flow of Data
A13
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

5- Conclusion:
In summary, incorporating AI into the architectural design process represents a paradigm
shift in which technology collaborates with human creativity. Architects benefit from
synergies by using AI as a tool to explore, optimize, and overcome design challenges in
dynamic and innovative environments. The result is a harmonious fusion of human ingenuity
and technical ability in the field of architectural creativity. Integrating artificial intelligence
(AI) into architectural design has led to many achievements, including:
- Assisting Architects with Data Processing: AI helps architects collect and process large
amounts of data, generate multiple variables, and extract thousands of unique models and
designs through architect-specified algorithms and inputs. However, the creative process
remains only in the realm of the human mind.
- Contribution of AI Algorithms to Design Interpretation: AI algorithms contribute to
interpreting the behavior patterns and preferences of architects through stored data,
producing unique and non-repetitive designs to create a customized experience for each
architect.
- Enhancing Efficiency Across Architectural Design stages: AI patterns and applications
can help architects complete various stages of the design process with desired efficiency,
saving time and effort from pre-design to design, construction, construction management,
and post-occupancy evaluation., save costs.
- AI Patterns Enhancing Efficiency in Architectural Design stages: While AI patterns
enhance the efficiency of architectural design phases, the continuous evaluation by design
process stakeholders (architects, owners, users, experts, etc.) is essential at decisions and
control are necessary at every stage of the project.
- Significant Role of AI in Architectural Design stages: AI plays a key role in all stages of
the architectural design process. However, it cannot serve as an assessment tool or decision-
making aid. These aspects remain the domain of the human mind.
- AI Will Not Replace Architects: AI will not replace architects but can be used as an
assisting tool throughout the design process.

6-Recommendations:
- Integrating AI learning into architecture curricula: It is recommended that AI learning and
the study and use of AI applications related to architectural design be integrated into the
curriculum of architecture universities and institutes of architecture in Egypt.
– Promoting a culture of AI use. It is essential to spread the culture of using AI as a tool in
the architectural design process to students and practitioners.
– Architects must learn and leverage new AI technologies to maintain their professional
reputation.
– Using AI in design firms allows design and engineering consulting firms to integrate AI
patterns and applications into the design process of each project to save time and effort and
avoid run-time errors., thereby significantly facilitating and controlling all design process
activities.
– It is essential to build an integrated AI strategy for architectural design by developing a
comprehensive strategy for the use of AI technology in architectural design through
complementary research that takes into account all relevant details.

A14
 Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16

Reference:
1- As, I., Pal, S., & Basu, P. (2018). Artificial intelligence in architecture: Generating conceptual
design via deep learning. International Journal of Architectural Computing, 16(4).
doi:https://doi.org/10.1177/1478077118800982
2- Borglund , C. (2022). Artificial Intelligence in Architecture and its Impact on Design Creativity
‘A Study on how Artificial Intelligence Affect Creativity in the Design Process. Stockholm,
Sweden: KTH ROYAL INSTITUATION OF TECHNOLOGY.
3- Campesato, O. (February 13, 2020). Artificial intelligence, machine learning, and deep learning.
Mercury Learning and Information. doi: ISBN, 1683924673, 9781683924678
4- Chaillou, S. (2019). AI+Architecture: Towards a New Approach. USA: Harvard Graduate School
of Design.
5- Cudzik, J., & Radziszewski, K. ( 2018, Oct.). Artificial Intelligence Aided Architectural Design,‖
AI FOR DESIGN AND BUILT ENVIRONMENT. eCAADe 36, 1, 77–84.
doi:https://www.researchgate.net/publication/328018944
6- Gallo, G., & wirz, f. (2020). The role of Artificial Intelligence in architectural design:
conversation with designers and researchers. Proceedings of S.Arch 2020, the 7th international
conference on architecture and built environment, (pp. 198-205). Tokyo, Japan. doi:ISBN 978-3-
9820
7- Haenlein, M., Kaplan, A., Tan , C.-W., & Zhang , P. (2019). Artificial intelligence (AI) and
management analytics. Journal of Management Analytics, 341-343.
doi:https://doi.org/10.1080/23270012.2019.1699876
8- Haenlein, M., & Kaplan, A. (2019). A Brief History of Artificial Intelligence: On the Past,
Present, and Future of Artificial Intelligence. California Management Review, 61(4), 5-14.
doi:https://doi.org/10.1177/0008125619864925
9- Khan, H. (2021, 09). Different Types of Artificial Intelligence Systems. Types of AI, p. 50.
doi:fossguru.com/types-of-ai-different-types-of-artificial-intelligence-systems
10- Navarathna, P. J., & Malagi, V. P. (2018, Dec. 13-14). Artificial Intelligence in Smart City
Analysis. International Conference on Smart Systems and Inventive Technology (ICSSIT), pp. 44-
47. doi:doi: 10.1109/ICSSIT.2018.8748476.
11- Rao, S. (2022, Apr. 6). 10 Examples of Artificial Intelligence in Construction. Retrieved 10 23,
2023, from The Benefits of AI in Construction: constructible.trimble.com › the-benefits-of-ai-
inconstruction
12- Walch , K. (2019, 9 17). The Seven Patterns Of AI. Retrieved 10 15, 2023, from
https://www.forbes.com.: https://www.forbes.com/sites/cognitiveworld
13- Wang, M., Cai , J., & Tian, T. (2018, Apr.). DEFINITION, APPLICATION AND INFLUENCE
OF ARTIFICIAL INTELLIGENCE ON DESIGN INDUSTRIES. Landscape Architecture
Frontiers, 6(2). doi: http://dx.doi.org/10.15302/J-LAF-20180207
14- wenjun, M., & malaeb, J. ( sep 2019). AIA Artificial Intelligence in Architecture GENERAL
UNDERSTANDING AND PROSPECTIVE STUDIES. China: SHANGHAI JIAO TONG
UNIVERSITY.
15- Zheng, H., & Yuan, P. (2021). A generative architectural and urban design method through
artificial neural networks. Building and Environment, 205.
doi:https://doi.org/10.1016/j.buildenv.2021.108178

A15
 ‫‪Nermen M.Matter /Et Al/Engineering Research Journal 181 (March2024) A1-A16‬‬

‫الذكاء االصطناعي في العمارة‪ :‬التكامل في عملية التصميم المعماري‬

‫ملخص‪ :‬فً اآلونة األخٌرة‪ ،‬تغلغل الذكاء االصطناعً )‪ (AI‬فً لطاعات متنوعة‪ ،‬وأحدث ثورة فً مختلف جونب الحٌاة‬
‫واستبدال البشر باآللة من خالل أتمتة المهام التً ٌؤدٌها البشر تملٌدًٌا‪ .‬وٌشمل ذلن أتمتة التصنٌع‪ ،‬وبعض طرق التعلٌم‪،‬‬
‫ودٌنامٌكٌات وسائل التواصل االجتماعً‪ .‬وعلى الرغم من هذه التطورات‪ ،‬إال أنه ال ٌزال هنان الكثٌر من المهام لم ٌستطع‬
‫ٌتفرد به اإلنسان‪ٌ .‬ستطٌع المعمارٌون االستفادة من‬
‫ّ‬ ‫الذكاء االصطناعً السٌطرة علٌها‪ :‬وعلى رأسها اإلبــــــــداع‪ ،‬الذي‬
‫تمنٌات الذكاء االصطناعً فً تسهٌل عملهم‪ ،‬وإنجاز المهام غٌر اإلبداعٌة بمساعدة تلن التمنٌات‪ ،‬وذلن لتوفٌر الولت والجهد‬
‫ثوان‪،‬‬
‫ٍ‬ ‫إلتمام العملٌات اإلبداعٌة من ابتكار وتطوٌر‪ .‬على الرغم من ثروة البٌانات التً ٌولدها الذكاء االصطناعً فً غضون‬
‫إال أنه ال تزال هذه العملٌة غٌر مألوفة ألغلب المعمارٌٌن والمصممٌن والمخططٌن‪ .‬وبدراسة الذكاء االصطناعً وأنماطه‪،‬‬
‫واجراءات عمله واتخاذه للمرارات‪ ،‬والتنبؤ بالنتائج والتطورات المرتبطة بالبٌانات واألحداث المعروضة علٌه؛ اتضح أنه‬
‫لد ٌُساهم بدور ف ّعال ومساعد للمعماري فً ممارسة أغلب مراحل عملٌة التصمٌم المعماري‪ .‬حٌث إنه ٌتمٌز بالتعلم المستمر‪،‬‬
‫مما ٌجعله ٌتطور ذاتٌا وٌحسن من أدائه‪.‬‬
‫ٌهدف هذا البحث إلى صٌاغة إطار لتمنٌن دور الذكاء االصطناعً وانماطه فً إعادة صٌاغة وممارسة مراحل عملٌة التصمٌم‬
‫المعماري‪ ،‬وإنجاز المهام غٌر اإلبداعٌة مما ٌوفر الكثٌر من الولت والجهد‪ .‬وٌساعد المعماري فً عملٌة االبداع وتطوٌر‬
‫افكاره‪ ،‬والوصول ألفضل النتائج وتمدٌم تصمٌمات مبتكرة‪.‬‬
‫ٌرتكز البحث على إطار نظري ٌسلط الضوء على األدبٌات ذات الصلة بالذكاء االصطناعً وأنواعه وأنماطه المتنوعة‪،‬‬
‫ودراسة مراحل العملٌة التصمٌمٌة فً ضوء الذكاء االصطناعً وحدوده وامكانٌاته وأثر ذلن على دور المعماري‪ ،‬حٌث‬
‫توصل البحث إلى وضع تصور ُممترح لتمنٌن دور الذكاء االصطناعً كمساعد للمعماري فً تأدٌة عمله‪ٌ ،‬عتمد فً الممام‬
‫األول على تطبٌك أنماط الذكاء االصطناعً المختلفة‪ -‬تبعا ألهدافها ولدراتها‪ ،‬وبالتالً توفٌر الدعم الشامل طوال عملٌة‬
‫التصمٌم المعماري‪.‬‬
‫الكلمات المفتاحية‪ :‬الذكاء االصطناعً – اإلبداع – عملٌة التصمٌم المعماري – أنماط الذكاء االصطناعً‬

‫‪A16‬‬

‫‪View publication stats‬‬