Progress and obstacles in the use of artificial intelligence in civil engineering: An in-

depth review
Ruchit Parekh 1, * and Olivia Mitchell 2
1 Department of Engineering Management, Hofstra University, New York, USA.
2 Department of Civil Engineering, University of California, Berkeley, USA.

International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

Publication history: Received on 11 August 2024; revised on 17 September 2024; accepted on 18 October 2024

Article DOI: https://doi.org/10.30574/ijsra.2024.13.1.1777

Abstract
Artificial Intelligence (AI) has emerged as a transformative force across various domains, with its potential to
revolutionize urban architecture gaining increasing recognition. This paper offers a detailed examination of AI's
application in the construction of public buildings, emphasizing its achievements, challenges, and future outlook. The
review spans all facets of civil engineering, including review processes, analysis, design, construction management,
geotechnical engineering, transportation planning, and construction oversight. AI methods, such as machine learning
and genetic algorithms, are employed in analysis and design to enhance processes, forecast material behavior, and
advance healthcare applications. In construction management, AI is utilized for project scheduling, resource
distribution, risk evaluation, and safety management. Geotechnical applications of AI provide precise soil property
estimation, soil damage assessment, and foundation construction improvements. Advanced technologies aid in
transportation planning, traffic prediction, intelligent transportation systems, and infrastructure enhancements.
Additionally, AI plays a crucial role in monitoring and maintaining public infrastructure, including bridge inspections,
pipeline integrity evaluations, and early defect detection through image processing and data analysis. Despite significant
advancements, challenges persist regarding AI's widespread adoption in civil engineering, including data availability,
AI model definitions, ethical issues, and the necessity for collaborative efforts. Addressing these challenges will require
the joint efforts of researchers, practitioners, and policymakers. Ultimately, AI's integration into civil engineering
demonstrates its potential to enhance the efficiency, safety, and sustainability of infrastructure systems. This review
summarizes the current knowledge, highlights challenges, and proposes directions for future research to advance AI
integration in civil engineering.

Keywords: Artificial Intelligence; AI Applications; Structural Analysis; Structural Design; Machine Learning

1. Introduction

1.1. Overview of AI in Civil Engineering

Artificial Intelligence (AI) has garnered significant attention recently for its potential to revolutionize various industries,
including civil engineering. AI technologies offer the promise of enhancing efficiency, accuracy, and safety across
numerous applications, such as analysis and design, project management, construction, geotechnical engineering,
transportation planning, and infrastructure maintenance.

Smith et al. (2018) highlighted that AI could streamline processes, forecast material behaviors, and enhance healthcare
within buildings. For instance, machine learning algorithms can sift through large datasets to uncover patterns that
refine design processes (Li et al., 2020). AI models can also assist in choosing appropriate construction materials by
analyzing historical data and predicting performance under various conditions (Chen et al., 2019).

* Corresponding author: Ruchit Parekh

Copyright © 2024 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0.
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

In construction management, AI demonstrates potential in project planning, resource allocation, risk evaluation, and
safety oversight. Wang and Liu (2020) explored AI algorithms to develop construction schedules, accounting for factors
such as labor, capacity, and project constraints. AI-driven risk assessment models are designed to identify and mitigate
potential construction risks (Li et al., 2021).

Geotechnical engineering benefits from AI applications as well. AI systems can predict soil behavior, map soil stability,
and enhance foundation design. Huang et al. (2019) employed AI algorithms to forecast soil strength from geotechnical
data, offering valuable insights for geotechnical engineers. AI methods are also used for earthquake hazard mapping,
which improves the assessment and management of seismically active areas (Yin et al., 2020) (Fig. 1).

AI contributes significantly to transportation planning through traffic forecasting, intelligent transportation systems,
and optimization. Chen and Xie (2020) developed traffic prediction models using AI to forecast traffic conditions and
refine traffic control strategies. Intelligent transportation integrates AI to boost traffic efficiency, safety, and security
(Huang et al., 2021).

In the realm of maintenance and infrastructure upkeep, AI has proven useful for tasks such as bridge inspections,
pipeline integrity assessments, and early detection of structural defects. For example, Li et al. (2020) proposed a
cognitive-based bridge analysis utilizing computer vision to examine images and identify potential issues. AI algorithms
are also employed to monitor pipeline integrity and detect leaks or damage (Kim et al., 2019).

Kaveh et al. (2023) introduced a meta-heuristic-based Artificial Neural Network (ANN) approach to analyze the plastic
boundaries of frames. This work combines metaheuristic algorithms with ANNs to enhance the accuracy of plastic limit
analysis, crucial for model design and evaluation. Similarly, Kaveh & Khavaninzadeh (2023) focused on predicting the
strength of fiber-reinforced polymers (FRP) using ANNs, employing four metaheuristics to train two ANNs,
demonstrating their effectiveness in FRP strength prediction.

Kaveh & Eskandari (2021) conducted a comparative study of different ANN architectures for analyzing two-story cradle
vaults, evaluating their performance and accuracy. Kaveh et al. (2018) investigated the shear strength of FRP reinforced
concrete beams using Group Data Management Process (GMDH) in neural networks, showcasing the GMDH-based
network’s accuracy in predicting shear strength.

Rofooei et al. (2011) explored neural networks to predict concrete slab deterioration over time and assess vulnerability
to seismic loads. This study illustrates how neural networks can forecast structural damage. Kaveh and Rahimi
Bondarabady (2004) proposed a method combining graphs, neural networks, and genetic algorithms to reduce
wavefronts in finite element analysis, highlighting its effectiveness in optimizing simulations in civil engineering
research.

In conclusion, AI holds considerable promise for transforming spatial design, development, project management, and
property maintenance within civil engineering. However, challenges such as data availability, AI model interpretation,
and ethical issues remain. Future research should address these challenges and facilitate the integration of AI into civil
engineering practices.

2. Significance of AI in Transforming the Field

The integration of AI into civil engineering holds significant promise for reshaping the field by enhancing traditional
practices and boosting efficiency, accuracy, and overall performance. This transformative potential is acknowledged by
both researchers and industry experts.

Bilal et al. (2020) noted that AI technology could streamline essential engineering processes, enhance design quality,
and refine decision-making. By leveraging AI algorithms and machine learning techniques, engineers can process vast
amounts of data, uncover patterns, and derive insights that were previously challenging to achieve. This capability
allows for more informed decision-making throughout the project lifecycle, leading to improved designs, cost
reductions, and better performance outcomes.

AI also plays a crucial role in project management and optimization within the construction industry. Tasks such as
design, scheduling, and resource allocation, which have traditionally been time-consuming and prone to human error,
could be transformed by AI. This shift would enable engineers to concentrate on the more complex and creative facets
of their work. Automation not only enhances productivity but also minimizes the risk of errors and rework, leading to
substantial time and cost savings (Wu et al., 2017).

1060
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

By utilizing historical data and AI models, engineers can forecast potential failures, pinpoint maintenance requirements,
and extend asset lifecycles. Transitioning from reactive rework strategies to proactive maintenance can enhance safety,
reduce downtime, and improve asset management (Kumar et al., 2021).

Sustainability is another critical aspect where AI can contribute significantly. Sustainable design and construction
practices are essential for minimizing environmental impacts and improving resource efficiency. AI can support these
objectives by optimizing material selection, reducing energy consumption, and enhancing waste management. For
instance, AI algorithms can evaluate building performance data to identify energy-saving opportunities, leading to the
development of energy-efficient models (Wibowo et al., 2020).

In summary, the role of AI in advancing civil engineering is evident in its ability to refine processes, automate routine
tasks, aid in predictive maintenance, and enhance sustainability. By harnessing AI's capabilities, engineers can
overcome traditional constraints and unlock new avenues for innovation, efficiency, and effectiveness in the built
environment.

3. Purpose and Scope of the Review Paper

The objective of this review paper is to offer a comprehensive overview of the application of AI in civil engineering,
emphasizing the advancements achieved, the challenges encountered, and future prospects.

This article aims to synthesize existing literature, identify major themes, and provide insights into the development of
expertise in civil engineering. The review encompasses all facets of civil engineering, including analysis and design,
construction management, geotechnical engineering, transportation planning, and infrastructure maintenance. Each
domain will be explored to understand the application of AI techniques, the outcomes realized, and the difficulties faced.
Specific AI applications within civil engineering will be introduced, detailing the methods, techniques, and algorithms
utilized in each area.

The paper will include case studies and examples to illustrate the practical implementation of AI in real-world civil
engineering scenarios. In addition to covering applications, the review will address the challenges and limitations
associated with AI in civil engineering. These challenges include issues related to data availability and quality, AI model
definition, ethical considerations, and the necessity for collaborative efforts.

Furthermore, the review will highlight areas for future research and potential development within the field of
construction. It will identify knowledge gaps, propose directions for further investigation, and suggest solutions to
existing problems.

Overall, this review aims to enhance understanding of AI applications in civil engineering, assess progress and
challenges, and offer valuable insights to researchers, practitioners, and policymakers. The goal is to contribute to the
ongoing exchange of knowledge, encourage further research and innovation, and facilitate the effective application of
AI in engineering practice.

4. Applications of AI in structural analysis and design

4.1. General
AI technology has significantly impacted the fields of structural analysis and design by transforming traditional
methodologies and enhancing engineers' capabilities to improve designs, predict material behaviors, and advance
structural health monitoring. Key applications of AI in this domain include:

 Optimization through Machine Learning: Machine learning algorithms, such as neural networks and genetic
algorithms, are utilized to enhance design processes. These algorithms analyze extensive datasets, evaluate
various design alternatives, and identify optimal solutions that meet performance criteria. For instance, genetic
algorithms have been employed to refine the shape and structure of truss systems, thereby enhancing their
efficiency (Shen et al., 2019; Li et al., 2020).
 Predictive Models for Product Behavior: AI-based predictive models are capable of forecasting the behavior of
materials under different conditions by learning from historical data. These models assist in material selection
by providing insights into how materials will perform in specific applications. Chen et al. (2019) developed a
neural network model to estimate the shear strength of sand-cement mixtures, offering valuable guidance for

1061
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

geotechnical applications.
 Structural Health Monitoring: AI plays a crucial role in structural health monitoring by enabling real-time
analysis and diagnosis of structural components. Data-driven methods, including machine learning and deep
learning, process sensor data to detect defects, predict potential damage, and assess structural integrity. Lee et
al. (2020) proposed an AI-based approach that uses computer vision techniques to analyze images for
identifying potential issues.
 Optimization of Structural Design: AI algorithms can optimize structural design by considering multiple design
parameters and constraints. These algorithms improve work patterns, reduce material usage, and enhance
cost-effectiveness by finding the most efficient design solutions. Wu et al. (2019) demonstrated AI-based
optimization in reducing the weight of steel truss bridges while maintaining design safety.
 Intelligent Decision Support Systems: AI enhances decision-making during the design process by analyzing
complex datasets and providing recommendations. Intelligent decision support systems incorporate design
requirements, constraints, and historical data to aid engineers in crafting optimal solutions (Dorigo et al., 2020).

5. Applications of Machine Learning Techniques for Structural Optimization

Machine learning techniques offer efficient and effective approaches for enhancing design and performance in
structural optimization. By utilizing large datasets and sophisticated algorithms, machine learning enables engineers to
explore complex designs, optimize processes, and improve overall efficiency. Key aspects of machine learning in
structural optimization include:

 Neural Networks for Structural Design: Neural networks, inspired by the human brain's neural connections,
are employed for optimizing structural designs. These models learn the relationship between design
parameters and performance outcomes, facilitating better design solutions. For instance, Li et al. (2020)
introduced a machine learning approach using neural networks to enhance the design of steel-concrete
composite beams, improving their performance.
 Genetic Algorithms for Layout Optimization: Genetic algorithms, which are based on evolutionary principles,
are extensively used for optimizing structural layouts. These algorithms mimic natural selection processes to
find optimal solutions by selecting, combining, and evolving design configurations. Shen et al. (2019) applied a
genetic optimization method to minimize the weight of lattice structures while meeting structural
requirements.
 Reinforcement Learning for Control Systems: Reinforcement learning, a branch of machine learning focused on
decision-making and control, is used to develop control strategies for complex systems. By training models to
make decisions based on feedback and rewards, reinforcement learning enhances performance management
in control systems. For example, Kim et al. (2019) created an intelligent pipeline monitoring system that utilizes
deep learning and reinforcement learning to enhance pipeline safety and efficiency.
 Support Vector Machines for Classification and Prediction: Support Vector Machines (SVMs) are popular
machine learning models for classification and prediction tasks. In engineering, SVMs are used to classify and
predict damage in structural components. Huang et al. (2019) demonstrated the use of SVM-based methods to
predict soil erosion, providing valuable insights for geotechnical applications.
 Particle Swarm Optimization with Machine Learning Ensemble: Particle Swarm Optimization (PSO) is a
population-based optimization method inspired by social behavior. Integrating machine learning techniques
with PSO enhances performance and speeds up convergence. Wang and Liu (2020) proposed a PSO algorithm
combined with cloud computing and machine learning to develop an efficient scheduling system, improving
project efficiency and resource allocation.

6. Applications of Predictive Models for Material Behavior

Predictive models utilizing AI are extensively employed in civil engineering to forecast the behavior of various materials
under different conditions. These models leverage historical data and advanced algorithms to enhance understanding
of material properties, aiding engineers in making informed decisions regarding material selection, construction, and
maintenance. Key applications of predictive models for material behavior include:

 Concrete Strength Estimation: Accurately estimating the compressive strength of concrete is crucial for design
and quality control. AI-based predictive models, such as neural networks and support vector machines, are
used to forecast the strength of concrete mixes and treatments. For example, Chen et al. (2019) developed a
neural network model to predict the shear strength of sand-cement mixtures, facilitating more accurate
predictions in geotechnical applications.

1062
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

 Steel Property Estimation: Predicting the properties of steel, such as yield strength, tensile strength, and
elasticity, is essential for design and engineering analysis. AI models, including neural networks and regression
models, have been employed to estimate steel properties based on factors like chemical composition, heat
treatment, and design parameters. Chen et al. (2020) proposed a neural network model to predict the tensile
strength of high-strength steels, showcasing the effectiveness of AI-based methods in property prediction.
 Soil Behavior Modeling: Predictive models for soil behavior are important for geotechnical projects. Techniques
such as machine learning and fuzzy logic are used to simulate complex soil behaviors, including compression,
shear strength, and consolidation. For instance, Cano et al. (2019) developed a machine learning model to
predict soil compaction, aiding in foundation design and soil analysis.
 Composite Material Analysis: Composite materials, such as Fiber Reinforced Polymers (FRP), exhibit unique
strength and durability characteristics. AI-based prediction models are used to analyze and optimize the
behavior of these composites. Machine learning algorithms can evaluate factors like composition, fiber
orientation, and manufacturing processes to predict composite properties and performance. Zhang et al. (2020)
created a predictive model to assist in the design and development of FRP beams by estimating their bending
behavior based on intermediate energy properties.
 Asphalt Pavement Performance Prediction: Forecasting the performance and deterioration of asphalt
pavements is crucial for maintenance and repair strategies. Advanced models, including regression and time-
series analysis, are used to predict pavement performance indicators such as cracking, rutting, and roughness,
taking into account factors like traffic load and environmental conditions. Sun et al. (2020) proposed a machine
learning-based approach to predict the cracking performance of asphalt pavements, contributing to improved
pavement quality management.

7. Structural Health Monitoring (SHM)

Structural Health Monitoring (SHM) is a crucial aspect of civil engineering aimed at ensuring the safety and integrity of
structures throughout their lifespan. AI technology has significantly enhanced SHM by providing advanced tools for
analyzing sensor data, detecting anomalies, and forecasting structural health. The integration of AI with SHM enables
real-time monitoring, precise diagnostics, and effective maintenance strategies. Key applications of AI in SHM include:

 Anomaly and Damage Detection: AI algorithms, including machine learning and deep learning, are employed to
analyze sensor data and detect anomalies that may signal damage or deterioration. These algorithms can learn
from historical data to recognize patterns of normal and abnormal behavior. For instance, Lee et al. (2020)
utilized AI to investigate damage patterns and anomalies, enhancing the accuracy of damage detection.
 Damage Identification and Classification: AI systems can identify and categorize various types of damage such
as cracking, corrosion, and fatigue. Machine learning algorithms like support vector machines and random
forests analyze sensor data to detect patterns indicative of different damage types. Hu et al. (2019) developed
a machine learning-based system for classifying damage in steel structures, focusing on differentiating among
various damage forms.
 Predicting Critical Lifespan: AI-based models can forecast the critical lifespan of structures by analyzing sensor
data and degradation trends. These models use machine learning and statistical methods to predict future
performance and behavior of structures. Liang et al. (2020) proposed a predictive method for estimating the
fatigue life of steel bridges, aiming to optimize maintenance and extend the service life of the structure.
 Data Fusion and Multi-Sensor Integration: AI technology facilitates the integration of data from multiple
sensors, enhancing the accuracy and reliability of SHM models. By combining information from various sensors
such as gauges and temperature sensors, AI algorithms provide a comprehensive view of structural health. Lee
et al. (2019) introduced an AI-driven data fusion framework for SHM that integrates data from different sensors
to improve detection and localization of structural issues.
 Real-Time Decision Support: AI algorithms enable real-time processing of sensor data to support decision-
making in healthcare. These algorithms analyze data trends and evaluate conditions to provide timely and
informed decisions regarding structural maintenance and safety.

8. Overview of AI

8.1. General
Artificial Intelligence (AI) is a field dedicated to creating intelligent systems capable of performing tasks that typically
require human cognitive functions. In the realm of analysis and design, AI is instrumental in enhancing efficiency,

1063
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

accuracy, and automation. This section provides an overview of AI, including fundamental concepts, tools, and
processes, as well as insights from key experts.

 AI Terminology and Definitions: AI is defined as the simulation of human intelligence in machines designed to
mimic human thought processes and decision-making. It encompasses various subfields such as machine
learning, natural language processing, computer vision, and expert systems. Russell and Norvig (2016) offer a
comprehensive overview of AI, discussing its historical evolution, foundational concepts, and applications.
 Machine Learning Algorithms: Machine learning is a subset of AI that focuses on enabling machines to learn
from data and improve their performance autonomously, without explicit programming. It involves developing
algorithms that recognize patterns, make predictions, and adapt to new data. Popular machine learning
algorithms include decision trees, support vector machines, random forests, and neural networks. Bishop
(2006) provides an in-depth introduction to these algorithms, covering both fundamental and advanced
techniques.
 Deep Learning and Neural Networks: Deep learning, a branch of machine learning, involves artificial neural
networks with multiple layers that learn hierarchical data representations. These deep neural networks are
adept at processing complex patterns and making precise predictions. Deep learning has significantly impacted
applications like image and speech recognition. Goodfellow et al. (2016) offer a comprehensive guide on deep
learning, covering various aspects of design, optimization, and implementation.
 Data Mining and Knowledge Discovery: Data mining is the process of extracting valuable patterns and insights
from large datasets using AI techniques such as machine learning and statistical analysis. It involves uncovering
relationships and trends within data to aid decision-making. Fayed et al. (1996) explore the principles,
methods, and challenges of data mining in their foundational work on knowledge discovery and decision
support.

This overview establishes the core concepts of AI, including machine learning algorithms, deep learning, and data
mining. These principles underpin numerous AI applications in analysis and design, which will be further examined in
the subsequent sections of this review article.

9. Definition and Components of AI

AI Definition: Artificial Intelligence (AI) refers to the development of computer systems capable of performing tasks
that typically require human intelligence. These tasks include learning, reasoning, problem-solving, and decision-
making. AI integrates various techniques and methods to enable machines to replicate or simulate human cognitive
functions.

9.1. Components of AI
 Machine Learning: Machine learning is a subset of AI focused on creating algorithms and models that allow
computers to learn from and improve upon data without explicit programming. These algorithms analyze
patterns, extract insights, and make predictions or decisions based on learned data. Key types of machine
learning include:
o Supervised Learning: Learning from labeled data to make predictions.
o Unsupervised Learning: Finding patterns or structures in unlabeled data.
o Reinforcement Learning: Learning through rewards and penalties to make decisions.
 Deep Learning and Neural Networks: Deep learning is a branch of machine learning that uses multi-layered
neural networks to model complex patterns and make decisions. These deep neural networks mimic the human
brain's structure, allowing them to tackle intricate tasks such as image and speech recognition. Deep learning
excels in recognizing patterns and features in large datasets.
 Natural Language Processing (NLP): NLP is a field of AI dedicated to enabling computers to understand,
interpret, and generate human language. It involves analyzing text or spoken information and performing tasks
such as:
o Translation: Converting text from one language to another.
o Sentiment Analysis: Determining the emotional tone of text.
o Text Classification: Categorizing text into predefined groups.
o Information Extraction: Identifying and extracting useful information from text.
 Computer Vision: Computer vision is an AI field focused on interpreting and analyzing visual data, such as
images and videos. Key tasks in computer vision include:

1064
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

o Object Recognition: Identifying objects within images.

o Image Segmentation: Dividing an image into meaningful parts.
o Tracking: Following the movement of objects over time.
o Scene Understanding: Understanding and interpreting the context of visual scenes.
 Information Representation and Visualization: This component involves capturing and storing information in
formats that machines can process and use for decision-making. It includes:
o Creation of Information Systems: Structuring data in a way that is usable by AI.
o Ontologies and Expert Systems: Developing frameworks for representing knowledge and reasoning.
 Planning and Decision Making: AI systems use algorithms and processes to develop plans, execute actions, or
make decisions based on available information. Techniques in this area include:
o Search Algorithms: Finding solutions to complex problems.
o Optimization Methods: Enhancing performance by adjusting parameters.
o Predictive Modeling: Forecasting future outcomes based on current data.
AI systems integrate these components to enable machines to learn, reason, and act intelligently. By combining these
technologies, AI can solve complex problems, adapt to new situations, and improve continuously based on data and
experience.

10. Machine Learning Algorithms

Machine learning algorithms are fundamental to AI systems, enabling computers to learn from data and make
predictions or decisions without explicit programming. Here is an overview of some key machine learning algorithms
and contributions from the authors who advanced them:

10.1. Linear Regression

 Description: Linear regression is a supervised learning algorithm used to predict a continuous target variable
based on one or more predictor variables. It models the relationship between the target and the predictors
using a linear equation. The goal is to find the coefficients of this equation that minimize the sum of squared
errors between the predicted and actual values.
 Key Contribution: Hoerl and Kennard (1970) extended the concept of linear regression to address
multicollinearity issues with techniques such as ridge regression.

10.2. Decision Trees

 Description: Decision trees are used to create models that make decisions based on a series of hierarchical
rules. The algorithm splits the data into subsets based on feature values, aiming to maximize information gain
or minimize impurity at each node.
 Key Contribution: Breiman et al. (1984) introduced the Classification and Regression Tree (CART) algorithm,
which forms the basis for many decision tree models.

10.3. Random Forest

 Description: Random forest is an ensemble learning method that combines multiple decision trees to improve
predictive performance. It builds a collection of decision trees using random subsets of data and features. The
final prediction is made by aggregating the predictions of all individual trees.
 Key Contribution: Breiman (2001) proposed the random forest algorithm, which has proven effective across
various applications by reducing overfitting and improving accuracy.

10.4. Support Vector Machines (SVM)

 Description: Support vector machines are supervised learning algorithms used for classification and regression
tasks. SVM aims to find the optimal hyperplane that separates data points of different classes with the maximum
margin. It employs kernel functions to map data into higher-dimensional spaces, making it easier to find a
separating hyperplane.
 Key Contribution: Vapnik (1995) introduced the SVM algorithm and provided theoretical foundations for its
use, including the concept of margin maximization.

10.5. Neural Networks

 Description: Neural networks, particularly deep learning models, have gained prominence for their ability to

1065
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

model complex patterns. A neural network consists of layers of interconnected neurons, where each neuron
processes input data, applies an activation function, and passes the output to the next layer. Deep learning
involves multiple hidden layers to capture intricate data representations.
 Key Contribution: Rumelhart et al. (1986) developed the backpropagation algorithm, which is widely used to
train neural networks by adjusting weights through gradient descent.

These algorithms form the backbone of many AI applications, enabling systems to learn from data, make predictions,
and adapt to new information. Each algorithm has its strengths and is suited to different types of problems, contributing
to the versatility and effectiveness of machine learning in various domains.

11. Deep Learning and Neural Networks

Deep learning, a subfield of AI, focuses on the use of artificial neural networks to process and analyze data. These
networks are inspired by the human brain's structure and functionality, particularly the connections between neurons.
Here’s an overview of key concepts and applications in deep learning:

11.1. Artificial Neural Networks (ANN)

 Description: Artificial neural networks are foundational to deep learning. They consist of interconnected
artificial neurons that process information similarly to the human brain. Each neuron receives input, applies an
activation function, and produces an output. Weights on connections between neurons determine the strength
of their interactions.
 Key Contribution: The architecture of ANNs allows for learning complex patterns in data through multiple
layers of interconnected neurons.

11.2. Deep Neural Networks (DNN)

 Description: A deep neural network is an ANN with multiple hidden layers between the input and output layers.
These layers enable DNNs to learn hierarchical representations of data, capturing more abstract features. This
depth allows for more sophisticated feature extraction and improves accuracy in tasks such as classification
and prediction.
 Key Contribution: DNNs excel at processing and learning from large datasets with intricate patterns, making
them powerful tools for various applications.

11.3. Convolutional Neural Networks (CNN)

 Description: Convolutional neural networks are specialized types of deep neural networks used primarily for
image and video analysis. CNNs employ convolutional layers to automatically and adaptively learn spatial
hierarchies of features. They reduce the number of parameters by using shared weights and pooling layers,
which helps in feature extraction.
 Key Contribution: CNNs have significantly advanced fields like image recognition, object detection, and image
segmentation, revolutionizing computer vision tasks.

11.4. Recurrent Neural Networks (RNN)

 Description: Recurrent neural networks are designed to process sequential data, such as time series or natural
language. RNNs use feedback loops to maintain information from previous steps, making them suitable for tasks
where context or temporal information is essential. Variants like Long Short-Term Memory (LSTM) and Gated
Recurrent Units (GRU) address limitations in standard RNNs, such as difficulty in learning long-term
dependencies.
 Key Contribution: RNNs and their variants are particularly effective in handling tasks like language modeling,
speech recognition, and time-series forecasting.

11.5. Applications of Deep Learning and Neural Networks

 Image Classification: Deep neural networks, particularly CNNs, have been highly successful in image
classification tasks. For example, Krizhevsky et al. (2012) demonstrated the effectiveness of CNNs in the
ImageNet Large-Scale Visual Recognition Challenge, setting new benchmarks in image recognition.
 Translation: Neural networks have also advanced machine translation. Sutskever et al. (2014) introduced a
sequence-to-sequence model that has become the foundation for modern neural machine translation systems,
enhancing translation accuracy and fluency.

1066
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

 Speech Recognition: Deep learning has improved speech recognition significantly. Graves et al. (2013)
developed the Connectionist Temporal Classification (CTC) method, which has been crucial for training neural
networks to recognize speech more effectively.
 Recommendations: Neural networks are used in recommendation systems to provide personalized
suggestions. For instance, the work by She et al. (2017) on neural network matrix factorization has contributed
to advanced recommendation algorithms.
Deep learning and neural networks are rapidly evolving fields, driving innovation across numerous domains and
advancing AI research and applications.

12. Data Mining and Knowledge Discovery

Data mining and knowledge discovery are crucial components of AI that involve extracting valuable information and
patterns from large datasets. In civil engineering, these techniques are applied to various fields, such as structural
analysis, construction, and housing planning, to uncover hidden patterns and relationships.

12.1. Mining Process

12.1.1. Association Rule Mining

 Description: This technique identifies relationships or patterns within datasets. It is often used to understand
how different variables interact with each other.
 Example: Elnimeiri et al. (2015) used association rule mining to analyze the impact of climate on construction
delays, revealing how weather conditions affect project timelines.

12.1.2. Classification and Regression

 Description: Classification and regression techniques predict and classify data based on historical patterns.
Classification assigns data to predefined categories, while regression estimates continuous values.
 Example: Dong et al. (2016) utilized decision trees, a classification technique, to predict slope stability, helping
in assessing potential landslides.

12.1.3. Cluster Analysis

 Description: Cluster analysis groups similar data points based on their characteristics. It helps in identifying
patterns and anomalies within the data.
 Example: Asprone et al. (2019) used cluster analysis to identify similar buildings, facilitating the
implementation of retrofitting strategies.

12.2. Information Analysis

 Description: This iterative process involves extracting insights from data through various stages: data
collection, specification selection, design, and result interpretation.
 Example: Luo et al. (2020) employed knowledge discovery techniques to explore factors affecting productivity
in construction, demonstrating how data can inform better construction practices and decision-making.

12.3. Big Data in Civil Engineering

 Description: With the rise of big data, advanced data mining techniques are essential for processing and
analyzing large and complex datasets. Big data analytics combines data mining, machine learning, and statistical
methods to extract insights from unstructured and large datasets.
 Example: Garg et al. (2021) used big data analytics to analyze urban traffic data for traffic forecasting, which
helps in managing and optimizing urban transportation systems.
 Key Takeaway: Data mining and knowledge discovery are vital for extracting actionable insights from large
datasets in civil engineering. These techniques support decision-making, predictive modeling, and
optimization, ultimately improving performance and efficiency across various domains of the field.

13. AI-Based Structural Analysis Techniques

AI-based structural analysis techniques leverage advanced AI methods to enhance the accuracy, efficiency, and
feasibility of infrastructure design. These technologies utilize machine learning algorithms, neural networks, and other

1067
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

AI methods to model, simulate, and analyze the behavior of structures, predict their performance, and identify potential
issues.

13.1. Neural Networks for Structural Analysis

 Description: Neural networks, inspired by the human brain's structure, are used to predict and simulate
structural responses. They learn from historical data to anticipate how structures will behave under various
conditions.
 Example: Jangid et al. (2015) applied artificial neural networks to predict the dynamic response of buildings to
earthquake excitation, demonstrating the capability of neural networks to model complex structural behaviors.

13.2. Genetic Algorithms for Optimizing Structural Analysis

 Description: Genetic algorithms, inspired by natural selection, are used to optimize design parameters in
structural analysis. These evolutionary optimization techniques help refine element sizes, material
distribution, and shape configurations.
 Example: Saka and Kumbasar (2016) utilized a genetic algorithm to optimize the design of steel trusses,
highlighting the effectiveness of genetic algorithms in improving structural designs.

13.3. Machine Learning for Structural Health Monitoring

 Description: Machine learning algorithms are employed to monitor structural health by detecting anomalies,
predicting structural behavior, and assessing the condition of structures. They analyze sensor data to identify
deviations and provide early warnings of potential failures.
 Example: Bhattacharya et al. (2019) used machine learning to detect real-time damage from vibration data,
showcasing how machine learning can enhance the monitoring and maintenance of structural health.

13.4. Computational Intelligence for Nonlinear Structural Analysis

 Description: Computational intelligence techniques, including fuzzy logic, genetic programming, and swarm
intelligence, are used to handle complex behaviors and uncertainties in nonlinear structural analysis. These
methods address various types of nonlinearity such as material, geometric, and dynamic nonlinearity.
 Example: Zhao et al. (2017) employed fuzzy logic-based computational intelligence to analyze nonlinear
models, demonstrating the application of advanced computational methods to tackle complex structural
problems.

14. AI-Based Finite Element Analysis

Finite Element Analysis (FEA) is a crucial mathematical method used in civil engineering to simulate and analyze the
behavior of structures under various conditions. The integration of AI into FEA has significantly enhanced its efficiency,
accuracy, and automation. Here’s an overview of key AI-based techniques in FEA:

14.1. Proxy Modeling and Response Surface Methods

 Description: Proxy modeling techniques, including neural networks, support vector machines, and Gaussian
processes, are used to predict structural behavior based on finite FEA simulations. These surrogate models help
estimate structural responses to different loading scenarios, reducing the computational cost associated with
repeated FEA simulations.
 Example: Nguyen et al. (2018) utilized artificial neural networks as a proxy model to estimate the ultimate
strength of reinforced concrete beams, demonstrating the efficiency of AI in predicting structural performance.

14.2. Optimization and Design

 Description: AI algorithms such as genetic algorithms, particle swarm optimization, and evolutionary
algorithms are integrated with FEA to enhance design optimization. These algorithms automatically search for
optimal designs, including dimensions, materials, and improvements, based on defined objectives and
constraints.
 Example: Deb et al. (2016) proposed a hybrid optimization method combining FEA and genetic algorithms to
improve the design of lattice structures, highlighting AI's role in optimizing structural designs.


1068
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

14.3. Machine Learning for Model Calibration

 Description: Machine learning algorithms are employed to calibrate finite element models by adjusting them to
align with experimental data or real-world observations. These methods improve prediction accuracy by
addressing uncertainties and discrepancies between models and actual behavior.
 Example: Azadbakht et al. (2019) used a hybrid machine learning approach to calibrate finite element models
for predicting the dynamic behavior of building structures, showcasing AI’s capability to refine model accuracy.

14.4. Deep Learning for Structural Analysis

 Description: Deep learning techniques, including convolutional neural networks (CNNs) and recurrent neural
networks (RNNs), have shown potential in various aspects of structural analysis. CNNs are effective for tasks
such as crack detection in concrete structures using images, while RNNs are utilized for time series analysis
and estimating model responses under dynamic loading conditions.
 Example: Li et al. (2021) developed a CNN-based approach for crack detection in concrete structures,
demonstrating the application of deep learning in structural health monitoring.

15. AI-Based Optimization Methods

AI-based optimization methods leverage advanced techniques and machine learning to enhance various aspects of
construction, including design, resource allocation, and job planning. These methods aim to improve efficiency, reduce
costs, and optimize performance. Here’s an overview of key AI-based optimization techniques:

15.1. Genetic Algorithm (GA)

 Description: Genetic algorithms are optimization techniques inspired by natural selection. They use
evolutionary processes to iteratively improve solutions based on a population of possible solutions. GAs are
applied in civil engineering to optimize design processes, material selection, and structural configurations.
 Example: Wu et al. (2016) utilized genetic algorithms to enhance steel cage design, considering multi-
performance objectives, demonstrating GA’s effectiveness in structural optimization.

15.2. Particle Swarm Optimization (PSO)

 Description: Particle swarm optimization simulates the social behavior of birds or fish to search for optimal
solutions. It involves a swarm of candidate solutions that adjust their positions based on their own experience
and the experience of their neighbors. PSO is used in civil engineering for optimizing resource allocation,
settlement patterns, and design parameters.
 Example: Hassanpour et al. (2018) applied PSO to improve the design of distribution systems, considering
multiple constraints and objectives, highlighting PSO’s versatility in engineering optimization.

15.3. Artificial Neural Networks (ANN)

 Description: Artificial neural networks are computational models inspired by the human brain’s structure and
function. ANNs are used for optimization tasks such as predicting system responses, optimizing concrete mix
designs, and estimating construction costs. They learn from data to make predictions and optimize processes.
 Example: Gandomi et al. (2015) used ANNs to optimize the mix design of self-compacting concrete,
demonstrating ANN’s application in optimizing material properties and construction processes.

15.4. Machine Learning-Based Optimization

 Description: Machine learning techniques, such as regression models, support vector machines, and random
forests, are combined with optimization algorithms to enhance performance. These hybrid systems are applied
in engineering tasks like scheduling, resource allocation, and process optimization.
 Example: Alavipanah et al. (2019) combined Support Vector Regression with Genetic Algorithms to optimize
construction scheduling, illustrating the integration of machine learning and optimization for effective
decision-making.

1069
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

16. AI-Based Structural Dynamics and Vibration Analysis

Structural dynamics and vibration analysis are crucial for understanding how architectural structures respond to
various forces. AI advancements are being applied to these fields to enhance dynamic analysis accuracy, detect vibration
patterns, and optimize structural design. Here’s an overview of AI-based methods in this area:

16.1. AI-Based Modal Analysis

 Description: Modal analysis involves studying the vibration characteristics of structures to understand their
dynamic behavior. AI algorithms are employed to extract modal parameters from vibration data, helping to
identify key vibration modes and structural behaviors.
 Example: Zhang et al. (2015) introduced a machine learning-based approach to identify structural defects from
vibration response measurements. Their method effectively used established machine learning techniques to
enhance defect detection and analysis.

16.2. Machine Learning for Damage Detection

 Description: Machine learning algorithms are utilized for detecting and identifying damage in structures by
analyzing vibration data. These algorithms are trained on data from both healthy and damaged structures to
learn patterns and anomalies.
 Example: Nguyen et al. (2018) applied a deep learning method using convolutional neural networks to identify
damage patterns from vibration data. This approach improved the accuracy of damage detection in structural
analysis.

16.3. AI-Based Optimization of Structural Design

 Description: AI algorithms, including genetic algorithms and particle swarm optimization, are used to optimize
structural designs based on dynamic performance requirements. These techniques help in refining design
parameters to enhance structural performance under dynamic loads.
 Example: Chu et al. (2019) proposed a hybrid optimization algorithm that combines cutting-edge technology
with artificial neural networks to optimize the design of tall buildings subjected to wind-induced vibrations.
This method demonstrates how AI can improve structural design efficiency and effectiveness.

16.4. AI-Based Structural Health Monitoring (SHM)

 Description: AI techniques are integrated into SHM systems to monitor and analyze the health of structures in
real time. Machine learning algorithms are used to detect deviations from normal behavior, providing early
warnings of potential issues.
 Example: Aravantinos et al. (2020) developed an SHM system that utilizes AI-based anomaly detection
algorithms for bridge monitoring. Their system effectively identifies structural issues and triggers alerts based
on vibration data analysis.

17. AI-Based Structural Design Approaches

AI-based structural design leverages advanced algorithms and techniques to enhance the efficiency, optimization, and
creativity in engineering design processes. By integrating machine learning, optimization algorithms, and generative
design methods, AI is revolutionizing how structural designs are developed and refined. Here’s a breakdown of various
AI-based structural design approaches:

17.1. Design Optimization and Efficiency

 Description: AI-based optimization processes aim to increase design efficiency by exploring the design space
more effectively. Optimization algorithms, combined with machine learning or surrogate modeling, help in
finding optimal design solutions.
 Example: Ma et al. (2018) applied genetic algorithms and artificial neural networks to optimize steel frame
designs, demonstrating how AI can enhance design efficiency and performance.

17.2. Generative Design and Evolutionary Algorithms

 Description: Generative design technology, combined with evolutionary algorithms, generates multiple design
options based on predefined goals and constraints. This approach fosters creative exploration of the design

1070
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

space and results in innovative solutions

 Example: Jin et al. (2019) utilized genetic algorithms in generative design to improve lattice structures,
showcasing how AI can create novel and efficient design solutions.

17.3. AI-Based Material Selection

 Description: Machine learning algorithms are used to predict and optimize material choices for specific designs.
This process evaluates factors such as strength, durability, and cost to select the most suitable materials.
 Example: Fasihi et al. (2017) used support vector machine models to predict the compressive strength of
materials for bridge construction, illustrating how AI can aid in material selection for structural applications.

17.4. Design Information Discovery and Decision Support Systems

 Description: AI-based methods for design information discovery extract valuable insights from historical data
and expert knowledge. Machine learning techniques and data mining algorithms are used to analyze design
data, identify patterns, and support decision-making.
 Example: Asprone et al. (2018) developed a decision-making framework using data mining to enhance the
design of intervention strategies, demonstrating the role of AI in informed design decisions.

17.5. AI-Based Structural Optimization

 Description: AI-based structural optimization uses machine learning and optimization algorithms to refine
structural designs for improved performance. This approach involves exploring alternative designs, evaluating
their effectiveness, and iterating to find the optimal solution.
 Example: Tabeshpour et al. (2021) proposed an AI-based optimization method for the design of steel trusses
using deep neural networks and genetic algorithms, highlighting how AI can optimize structural performance.

18. AI-Based Design Optimization and Parameterization

AI-based design optimization and parameterization use advanced AI techniques to enhance the process of optimizing
civil engineering design parameters. These strategies employ machine learning, evolutionary algorithms, and other AI
methods to explore the design space, identify effective solutions, and improve system performance. Here’s an overview
of these approaches:

18.1. Evolutionary Algorithms

 Description: Evolutionary algorithms such as genetic algorithms (GA), particle swarm optimization (PSO), and
evolutionary programming mimic natural selection processes to search for optimal solutions. These algorithms
iteratively evolve solutions to find the best design parameters.
 Example: Bakhshpoori et al. (2016) used genetic algorithms to optimize the design of steel–concrete composite
beams, demonstrating how evolutionary algorithms can refine structural design.

18.2. Optimization Based on Machine Learning

 Description: Machine learning techniques, including neural networks and support vector regression (SVR), are
applied to design optimization. These techniques learn from historical data to create predictive models that
optimize design parameters.
 Example: Hagen et al. (2017) optimized the design of reinforced concrete beams using artificial neural
networks, illustrating how machine learning can enhance design accuracy and efficiency.

18.3. Multi-Objective Optimization

 Description: Multi-objective optimization involves addressing several conflicting goals, such as cost, efficiency,
and sustainability, simultaneously. Evolutionary algorithms like the Non-Dominated Sorting Genetic Algorithm
(NSGA) are used to tackle these complex optimization problems.
 Example: Shahbazpour et al. (2018) used a multi-objective genetic algorithm to optimize the design of steel
moment frames, showing how AI can balance multiple objectives in structural design.

18.4. Shape and Topology Optimization

 Description: AI techniques are employed in shape and topology optimization to find the most effective spatial

1071
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

distribution of materials within a design space. This approach leads to innovative and efficient designs.
 Example: Zhang et al. (2020) applied generative adversarial networks (GANs) for shape optimization of lattice
structures, highlighting how AI can generate new and effective design solutions.

18.5. Design Space Parameterization

 Description: AI-based parameterization involves defining and representing the design space using intelligent
techniques, which facilitates the exploration of various design options.
 Example: Chen et al. (2021) used particle swarm optimization to parameterize bridge engineering design space,
demonstrating how parameterization can streamline design exploration.

19. AI-Based Generative Design and Evolutionary Algorithms

AI-based generative design and evolutionary algorithms have revolutionized architectural innovation and optimization.
These approaches leverage AI algorithms, optimization techniques, and computational modeling to explore extensive
design spaces, create diverse designs, and identify optimal solutions.

19.1. Generative Design

 Description: Generative design employs AI algorithms to explore a broad range of design possibilities based on
specified conditions, goals, constraints, and limitations. This method can lead to innovative and effective design
solutions in various fields, including civil engineering, architecture, and urban planning.
 Example: Mendez et al. (2019) utilized generative design processes to optimize interior structures for energy
efficiency, demonstrating how this approach can enhance building performance.

19.2. Evolutionary Algorithms

 Description: Evolutionary algorithms, inspired by natural evolution, use genetic operators such as mutation,
crossover, and selection to develop and evolve design solutions. These algorithms explore the design space to
identify optimal or near-optimal solutions and are applied in structural optimization, design, and site planning.
 Example: Feng et al. (2020) applied genetic algorithms to optimize bridge foot designs, improving structural
performance through evolutionary optimization techniques.

19.3. AI-Driven Design Discovery

 Description: AI-driven design discovery combines design transformation algorithms with machine learning and
AI to explore and evaluate multiple design options. This approach aims to uncover new and improved solutions
through computational models, simulations, and optimization.
 Example: Zhu et al. (2017) used AI-driven design search to enhance the performance of a pedestrian bridge,
showcasing the potential of AI in refining design aesthetics and functionality.

19.4. Optimizing Building Design

 Description: AI-based generative design and evolutionary algorithms are used to optimize building layouts,
focusing on improving energy efficiency, performance, and occupant comfort. By evaluating various design
variables, these methods can meet multiple goals and design criteria.
 Example: Huang et al. (2020) employed an AI-based generative design approach to optimize the energy
performance layout of office buildings, highlighting the role of AI in achieving sustainable building design.

19.5. Urban Planning and Infrastructure

 Description: AI-driven generative design and evolutionary algorithms are also applied to urban planning and
infrastructure. These techniques help explore alternative urban layouts, enhance transportation links, and
improve construction methods.
 Example: Shan et al. (2018) optimized the layout of bike-sharing stations in cities using adaptive systems,
improving accessibility and usability through AI-based design strategies.

20. AI-Based Material Selection

AI-based material selection utilizes advanced AI techniques to identify and optimize materials for various engineering
applications. By leveraging machine learning, data-driven approaches, and optimization techniques, AI helps analyze

1072
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

materials, operational requirements, environmental factors, and cost considerations to recommend the best products
for a given project.

20.1. Machine Learning for Material Selection

 Description: Machine learning algorithms analyze large datasets to gain insights into material performance and
suitability. These algorithms identify patterns and relationships between components, performance metrics,
and application requirements.
 Example: Wu et al. (2017) applied machine learning techniques, such as support vector regression and random
forests, to develop models for selecting materials in concrete construction, optimizing material choices based
on various performance criteria.

20.2. Data-Driven Product Selection

 Description: A data-driven approach involves using big data and statistical analysis to inform material selection
decisions. This approach integrates feature information, operational data, and background context to create
models that identify the best products based on project needs.
 Example: Li et al. (2019) used a data-driven approach to develop a material selection process for offshore wind
turbine support structures, utilizing data exploration and analysis to choose suitable materials for challenging
environments.

20.3. Multi-Criteria Decision Making

 Description: Material selection often requires evaluating multiple factors such as mechanical properties,
durability, sustainability, and cost. AI-based methods like fuzzy logic and the Analytical Hierarchy Process
(AHP) help in making these complex decisions by weighing different criteria.
 Example: Zahedi et al. (2020) used the Fuzzy AHP method to prioritize building materials, aiding decision-
making in bridge material selection by balancing various factors and preferences.

20.4. Data Generation

 Description: AI can generate new data with desirable properties using techniques like genetic algorithms and
neural networks. This process explores potential material compositions and manufacturing methods to
recommend solutions that meet specific needs.
 Example: Lin et al. (2018) utilized a neural network-based genetic algorithm to design high-performance
concrete, enhancing strength and durability through optimized material formulations.

20.5. AI-Based Materials Prediction

 Description: AI techniques, including machine learning and molecular dynamics simulations, predict material
properties by analyzing available data on materials and molecular structures. These predictions help in
selecting materials with desired electrical, thermal, and chemical properties.
 Example: Sun et al. (2019) used machine learning algorithms to predict the strength of cement products,
demonstrating how AI can forecast material performance based on historical and experimental data.

21. Benefits and Limitations of AI in Structural Analysis and Design

The implementation of AI in structural analysis and design offers numerous advantages, although there are also some
limitations that need to be considered. Understanding both the benefits and drawbacks is essential for effectively
leveraging AI in engineering projects.

21.1. Benefits

21.1.1. Improved Efficiency and Accuracy

 AI algorithms excel at processing large volumes of data and performing complex calculations with precision.
This significantly reduces the time and effort required for analysis and design tasks. AI tools can automate
repetitive functions such as modeling, optimization, and design, leading to greater efficiency. Additionally, AI
enhances the accuracy of predictive models, generating more reliable results.
 Example: Zhang et al. (2020) demonstrated that AI-based optimization algorithms improved the performance
and accuracy of design processes.

1073
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

21.1.2. Enhanced Model Performance and Safety

 AI technologies enable more efficient analysis and optimization, which can lead to improved model
performance and structural safety. AI-based algorithms can optimize designs by reducing material usage
without compromising structural integrity. Moreover, AI assists in integrating real-time data from structural
health monitoring systems, allowing for continuous assessment and detection of potential issues.
 Example: Liu et al. (2018) used AI to optimize the design of wind turbine towers, resulting in better
performance and increased safety standards.

21.1.3. Innovative Design Discovery

 AI-based generative design techniques foster creativity by allowing engineers to explore a wide range of design
possibilities. Machine learning and evolutionary algorithms can generate numerous design options and
evaluate their performance against conventional methods. This process encourages designers to find unique
solutions and challenge traditional approaches.
 Example: Bhoopalam et al. (2021) utilized AI-driven generative design to discover innovative solutions for
bridge design.

In summary, AI brings significant benefits to structural analysis and design, improving efficiency, accuracy, and
innovation. However, a balanced understanding of its capabilities and limitations is vital for its successful application
in engineering.

22. Limitations of AI in Structural Analysis and Design

While AI offers numerous benefits in structural analysis and design, several limitations need to be addressed for its
effective implementation:

22.1. Data Needs and Quality

 AI algorithms depend heavily on high-quality data for training and decision-making. Inadequate or biased data
can lead to inaccurate predictions and suboptimal results. Data availability might be limited, particularly for
specific or rare engineering scenarios. Additionally, issues related to data privacy and security must be
managed, especially when dealing with external or sensitive project data.

22.2. Interpretation and Explanation

 Many AI tools, particularly deep learning neural networks, often operate as "black boxes" with limited
interpretability. This lack of transparency can be problematic in engineering analysis and design, where
understanding the rationale behind design decisions or analysis results is crucial. Research is ongoing to
improve the interpretability of AI models to provide clearer insights into their decision-making processes.

22.3. Advancement Skills and Knowledge

 Effective use of AI in engineering requires expertise in both AI methods and civil engineering. Developing,
validating, and evaluating AI tools and algorithms necessitates specialized knowledge. Engineers must be
trained to use AI tools effectively and interpret their results accurately to ensure their proper application in
engineering contexts.

22.4. Data Requirements

 Artificial Neural Networks (ANNs) often need large volumes of high-quality data for training. Obtaining
sufficient and reliable data, especially for specialized or rare cases, can be challenging. Limited data can affect
the performance and generalization capabilities of ANNs. The "black box" nature of ANNs can also make it
difficult to understand the underlying conditions and variables influencing predictions.

22.5. Overfitting and Generalization

 ANNs are susceptible to overfitting, where the model becomes too tailored to the training data and performs
poorly on new, unseen data. To mitigate overfitting, techniques such as regularization and cross-validation
should be used. Ensuring that ANN models can generalize effectively to new and unpredictable scenarios is
crucial for their reliability.

1074
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

22.6. Budget and Time

 Training complex ANN models can be resource-intensive, both in terms of cost and time. Large-scale datasets
and complex architectures require significant computational resources for training and optimization. These
demands can limit the feasibility of using ANNs in environments with constrained budgets or timeframes.

22.7. Model Uncertainty and Confidence

 ANNs may struggle to provide measures of uncertainty or confidence in their predictions. In structural analysis
and design, understanding the reliability of forecasts is important for making informed decisions. The absence
of uncertainty estimates can be a limitation in applications where measurement uncertainty plays a critical
role.

Addressing these limitations is essential to fully harness the potential of AI in structural analysis and design.
Overcoming these challenges can lead to more efficient, accurate, and innovative engineering practices.

23. Future Directions and Emerging Trends in AI for Civil Engineering

As AI technology continues to evolve, several emerging trends and future directions are poised to significantly impact
the field of civil engineering. These advancements have the potential to enhance efficiency, sustainability, and overall
performance in construction and infrastructure projects:

23.1. Integration of AI and Building Information Modeling (BIM)

 BIM involves creating and managing digital representations of physical and functional characteristics of
buildings and infrastructure. The integration of AI with BIM can enhance the capabilities of digital models,
enabling advanced analysis, optimization, and decision-making throughout a project's lifecycle. AI can leverage
BIM data to improve design accuracy, construction processes, and operational efficiency.

23.2. Explainable AI and Process Models

 Explainable AI (XAI) focuses on making AI models and algorithms more transparent and understandable. In
architecture and engineering, providing explanations for AI-driven decisions is crucial for trust and
accountability. Developing AI models that can clearly articulate their reasoning will help engineers better
understand and validate AI-generated results.

23.3. Hybrid Approaches and AI-Powered Collaboration in Design

 The future of design will likely involve a hybrid approach that combines human expertise with AI capabilities.
This collaborative process can harness the strengths of both human creativity and AI efficiency. Engineers will
provide domain knowledge and critical thinking, while AI will assist with design generation, simulation, and
performance analysis.

23.4. Integration of AI into Manufacturing and Robotics

 AI integration into manufacturing and robotics holds promise for advancing automation, enhancing
construction processes, and improving safety. AI-enabled robots can undertake tasks such as constructing
walls, pouring concrete, and performing field research with increased precision and efficiency. Machine
learning can optimize equipment usage and reduce operational downtime.

23.5. AI for Sustainable Infrastructure

 With growing emphasis on sustainability, AI can play a pivotal role in enhancing energy efficiency, minimizing
waste, and improving the environmental impact of construction processes. AI technologies can support the
development of energy-efficient buildings, optimize transportation infrastructure, and manage water
resources more effectively.

23.6. Enhanced Sensing and Data Integration

 The proliferation of the Internet of Things (IoT) and advancements in information technology are generating
vast amounts of data. AI can process and analyze this complex data for real-time monitoring, predictive
maintenance, and risk assessment. Integrating data from various sensors and sources can provide deeper

1075
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

insights into infrastructure behavior and support better decision-making.

23.7. Ethical Decision-Making and AI Responsibility

 As AI becomes more embedded in civil engineering, addressing ethical considerations is crucial. This includes
managing issues related to bias, fairness, privacy, and security within AI systems. Developing ethical and
regulatory frameworks will be important to ensure responsible use of AI, enhance transparency and
accountability, and safeguard stakeholder interests.
By focusing on these future directions and emerging trends, the civil engineering sector can leverage AI to drive
innovation, improve efficiency, and foster sustainable development. Addressing challenges related to data quality,
model interpretability, and ethical considerations will be key to realizing the full potential of AI in the field.

24. Conclusion
In this review article, we explored the application of AI in model analysis and design within civil engineering. The key
findings are summarized as follows:

 Enhanced Accuracy and Efficiency: AI-based strategies have demonstrated significant improvements in the
accuracy, efficiency, and effectiveness of analysis and design procedures. These advancements lead to more
reliable and optimized engineering solutions.
 Real-Time Structural Health Monitoring: AI enables real-time monitoring of structural health, facilitating early
detection of potential issues and allowing for timely maintenance. This contributes to improved safety, stability,
and overall quality of structures.
 Optimization and Performance Improvement: Machine learning and optimization algorithms have been
effectively employed to enhance engineering processes, resulting in better performance, increased efficiency,
and optimal resource utilization.
 Innovative Design Methods: AI-driven design approaches, including modular and adaptive systems, have
proven successful in generating innovative and refined solutions, pushing the boundaries of traditional design
practices.
 Data-Driven Insights and Predictive Modeling: Data mining and knowledge discovery techniques have been
utilized to extract valuable insights from large datasets. These methods support informed decision-making and
contribute to accurate predictive modeling.
Overall, the integration of AI in civil engineering offers transformative potential, driving advancements in analysis,
design, and overall infrastructure management. The ability to leverage AI for improved performance, safety, and
innovation marks a significant step forward in the field.

24.1. Implications for Future Research and Practice

Based on the findings from this review, several implications for future research and practice in AI applications for
structural analysis and design are identified:

 Integration with Emerging Technologies: There is a need to combine AI with new technologies such as Building
Information Modeling (BIM), the Internet of Things (IoT), and cloud computing. This integration can enhance
collaboration, knowledge sharing, and coordination in construction projects, leading to more efficient and
effective practices.
 Improving Transparency and Trust: Future research should focus on AI methods that improve transparency
and interpretability of AI-generated analysis and designs. Building trust and acceptance among stakeholders is
crucial, and developing AI systems that provide clear explanations and rationale will support this goal.
 Development of Ethical Guidelines: Collaboration among researchers, industry experts, and regulators is
essential to develop standards and guidelines for the ethical and responsible use of AI in structural analysis and
design. Ensuring that AI applications are used responsibly and ethically will be key to their successful adoption.
 Advancements in Real-Time Monitoring and Maintenance: Continued research and development in AI
algorithms for real-time health monitoring, early warning systems, and predictive maintenance strategies are
necessary. These advancements will help ensure the safety and resilience of structures, making maintenance
practices more proactive and effective.
In conclusion, the application of AI in structural analysis and design has the potential to significantly transform civil
engineering practices. The review highlights the numerous advantages and opportunities AI offers, including improved
accuracy, efficiency, and innovation in design processes. However, further research, collaboration, and innovation are
required to fully realize AI’s potential and address challenges impacting its successful implementation.

1076
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

Compliance with ethical standards

Disclosure of conflict of interest

No conflict of interest to be disclosed.

References
[1] Alavipanah, S. K., Minaei, B., Tavakkoli-Moghaddam, R., & Hashemi, H. (2019). A hybrid SVM-GA approach for
optimizing construction project scheduling under uncertainty. Automation in Construction, 99, 221–231.
[2] Aravantinos, V., Chatzi, E. N., Papadimitriou, C., & Ntotsios, E. (2020). A novel AI-based methodology for vibration-
based structural health monitoring. Mechanical Systems and Signal Processing, 135, 2.
[3] Asprone, D., Di Ludovico, M., Fumo, G., Prota, A., & Manfredi, G. (2018). Discovering knowledge in structural
design processes: A data mining approach. Journal of Computing in Civil Engineering, 32, 3.
[4] Asprone, D., Prota, A., Manfredi, G., & Cosenza, E. (2019). Building clustering for optimal urban resilience
enhancement. Journal of Computing in Civil Engineering, 33, 3.
[5] Azadbakht, M., Seyedpoor, S. M., & Ostadan, F. (2019). Application of a hybrid intelligent system in calibration of
finite element models of building structures for earthquake-induced response prediction. Engineering Structures,
180, 69–83.
[6] Bakhshpoori, T., Kheyroddin, A., & Sobhani, J. (2016). Optimization of steel-concrete composite beams using a
genetic algorithm. Journal of Constructional Steel Research, 119, 1–10.
[7] Bhattacharya, S., Chowdhury, A., & Basu, B. (2019). Structural damage detection using machine learning with
application to a benchmark problem. Journal of Sound and Vibration, 455, 318–336.
[8] Bhoopalam, A., Patel, D., & Bhaskar, A. (2021). Generative design approach for low-cost pedestrian bridge: An
artificial intelligencebased structural optimization. Structures, 33, 1900–1913.
[9] Bilal, M., Farooq, S., Younas, M., & Ali, M. (2020). Role of artificial intelligence in civil engineering: A systematic
literature review. Advances in Civil Engineering, 2020, 1–14.
[10] Bishop, C. M. (2006). Pattern Recognition and Machine Learning. Springer.
[11] Brachman, R. J., & Levesque, H. J. (2004). Knowledge Representation and Reasoning. Elsevier.
[12] Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5–32.
[13] Breiman, L., Friedman, J. H., Olshen, R. A., & Stone, C. J. (1984).
[14] Classification and Regression Trees. Wadsworth International Group.Cano, M., Ghorbanizadeh, O., & Guédon, S.
(2019). Soft soil consolidation settlement prediction with machine learning methods. Geotechnical and Geological
Engineering, 37(6), 5281–5296.
[15] Chen, W., Lin, S., Wang, D., Zhang, Y., & Cao, S. (2021). An optimization algorithm for parameterization of design
space in bridge engineering. Journal of Computing in Civil Engineering, 35, 1.
[16] Chen, Y., Jin, W., Li, L., Zhou, Q., & Li, J. (2020). Prediction of tensile strength for high-strength steel using a neural
network. Advances in Mechanical Engineering, 12, 10.
[17] Chen, Y., & Xie, Y. (2020). Traffic prediction using deep learning: A generative adversarial network approach.
IEEE Transactions on Intelligent Transportation Systems, 21(8), 3528–3539.
[18] Chen, Z., Zhang, Y., Li, J., & Zhu, W. (2019). An artificial neural network approach for predicting the strength
parameters of sandcement mixtures. Neural Computing and Applications, 31(4), 1293–1303.
[19] Chu, X., Sun, Y., Jin, Z., & Sun, X. (2019). Optimal design of tall buildings under wind-induced vibrations based on
hybrid optimization algorithm. Engineering Structures, 198, 2.
[20] Deb, K., Karthik, S., & Subrahmanya, S. (2016). Design optimization of truss structures using finite element
analysis and genetic algorithm. Journal of Computing in Civil Engineering, 30, 6.
[21] Dong, P., Hou, D., Chen, X., & Yu, H. (2016). Application of decision tree in slope stability analysis. Advances in Civil
Engineering, 20, 16.

1077
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

[22] Dorigo, M., Poretti, S., Roa-Sepúlveda, C. A., & Tormos, P. (2020). Multiobjective optimization in civil engineering:
A systematic review. Archives of Computational Methods in Engineering, 27(4), 1399–1447.
[23] Elnimeiri, M., Al-Khaleefi, K., & Elnimeiri, M. (2015). Data mining of construction delay causes. Journal of
Construction Engineering and Management, 141, 1.
[24] Fasihi, M., Malek, S., & Zolfaghari, A. (2017). Concrete compressive strength prediction using machine learning
models. Journal of Computing in Civil Engineering, 31, 6.
[25] Fayyad, U., Piatetsky-Shapiro, G., & Smyth, P. (1996). From data mining to knowledge discovery in databases. AI
Magazine, 17(3), 37–54.
[26] Feng, Y., Yuan, X., & Wang, Z. (2020). Multi-objective optimization design of bridge piers based on genetic
algorithm. Journal of Bridge Engineering, 25, 1.
[27] Gandomi, A. H., Alavi, A. H., & Mirhosseini, R. (2015). Multi-objective self-compacting concrete mix design using
ANN and NSGA-II. Neural Computing and Applications, 26(4), 811–822.
[28] Garg, A., Rana, P., & Aggarwal, S. (2021). Urban transportation systems: Big data analytics and intelligent
transport systems. Sustainable Intelligent Transportation Systems (pp. 47–66). Springer.
[29] Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.
[30] Graves, A., Mohamed, A. R., & Hinton, G. (2013). Speech recognition with deep recurrent neural networks. In
Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference (pp. 6645–6649).
[31] Hagan, T. D., Nimbalkar, S. U., & McCormick, J. P. (2017). Neural network-based optimization of reinforced
concrete beams. Journal of Computing in Civil Engineering, 31, 5.
[32] Hassanpour, H., Saffar, B., & Asgarian, B. (2018). Multi-objective particle swarm optimization for resource-
constrained project scheduling problem. Journal of Construction Engineering and Management, 144, 9.
[33] He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T. S. (2017). Neural collaborative filtering. In Proceedings of the
26th International Conference on World Wide Web (pp. 173–182).
[34] Hoerl, A. E., & Kennard, R. W. (1970). Ridge regression: Biased estimation for nonorthogonal problems.
Technometrics, 12(1), 55–67.
[35] Hu, Z., Li, J., & Zhang, Y. (2019). A machine learning-based approach for steel structure damage classification.
Structural Control and Health Monitoring, 26, 11.
[36] Huang, L., Han, D., & Wang, D. (2019). Landslide susceptibility mapping based on extreme learning machine with
particle swarm optimization. Engineering Geology, 261, 2.
[37] Huang, W., Xu, Z., Liu, Z., & Xu, X. (2021). An intelligent transportation system for smart cities: A systematic review
and metaanalysis. Sustainable Cities and Society, 68, 2.
[38] Huang, Y., Qi, X., Wang, W., & Xie, X. (2020). An AI-based generative design approach for energy-efficient building
layout. Automation in Construction, 117, 2.
[39] Jangid, R. S., Chaudhari, H. B., & Trikande, N. J. (2015). Artificial neural network based design response prediction
of buildings under earthquake excitation. Journal of Civil Structural Health Monitoring, 5(2), 155–166.
[40] Jin, Y., Guo, Y., He, J., & Cheng, G. (2019). Optimization of truss structures using generative design and genetic
algorithm. Advances in Civil Engineering, 20, 19.
[41] Jurafsky, D., & Martin, J. H. (2019). Speech and Language Processing. Pearson.
[42] Kaveh, A., Bakhshpoori, T., & Hamze-Ziabari, S. M. (2018). GMDHbased prediction of shear strength of FRP-RC
beams with and without stirrups. Computers and concrete is an International Journal, 22(2), 197–207.
[43] Kaveh, A., & Eskandari, A. (2021). Analysis of double-layer barrel vaults using different neural networks; a
comparative study. International Journal of Optimization in Civil Engineering, 11(1), 113–141.
[44] Kaveh, A., & Khavaninzadeh, N. (2023). Efficient training of two ANNs using four meta-heuristic algorithms for
predicting the FRP strength. In Structures, 52, 256–272.
[45] Kaveh, A., & Rahimi Bondarabady, H. A. (2004). Wavefront reduction using graphs, neural networks and genetic
algorithm. International Journal for Numerical Methods in Engineering, 60, 1803–1815.

1078
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

[46] Kaveh, A., Seddighian, M. R., & Farsi, N. (2023). A metaheuristicbased artificial neural network for plastic limit
analysis of frames. International Journal of Optimization in Civil Engineering, 13(2), 143–154.
[47] Kim, S., Oh, H., Jang, J., Kim, J., & Kang, Y. (2019). Intelligent monitoring system for gas pipeline using deep learning
and reinforcement learning. Sensors, 19, 22.
[48] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet classification with deep convolutional neural
networks. In Advances in Neural Information Processing Systems (pp. 1097–1105).
[49] Kumar, R., Kassim, K. A., & Ahmad, N. (2021). Intelligent structural health monitoring of civil infrastructure: A
state-of-the-art review. Engineering Structures, 243, 2.
[50] Li, H., Ma, Y., & Wang, C. M. (2021). Automated detection of crack patterns in concrete structures using
convolutional neural networks. Journal of Computing in Civil Engineering, 35, 2.
[51] Li, M., Yang, S., Yu, T., & Zhou, M. (2020). An intelligent bridge health monitoring system based on deep learning.
Journal of Computing in Civil Engineering, 34, 2.
[52] Li, M., Yang, S., Yu, T., & Zhou, M. (2020). A machine learning framework for optimizing design of steel-concrete
composite beams. Journal of Construction Engineering and Management, 146, 1.
[53] Li, X., Gao, X., & Cheng, L. (2019). Data fusion in structural health monitoring using deep learning. Structural
Control and Health Monitoring, 26, 6.
[54] Li, X., Wang, Y., Yuan, J., Yu, Y., & Hao, H. (2019). Data-driven material selection for offshore wind turbine support
structures. Journal of Construction Engineering and Management, 145, 6.
[55] Li, Y., Liu, Q., Zhou, J., & Zhou, X. (2021). Artificial intelligence in construction safety management: State-of-the-
art and future trends. Journal of Construction Engineering and Management, 147, 1.
[56] Liang, W., Zhang, Z., Zhang, H., Li, Y., & Zhang, Y. (2020). Remaining fatigue life prediction of steel bridges using
deep learning. Engineering Structures, 213, 2.
[57] Lin, Y., Xiao, J., Yao, Y., & Zhang, Y. (2018). Design of high-performance concrete mixtures using a genetic
algorithm-based neural network approach. Journal of Materials in Civil Engineering, 30, 3.
[58] Liu, J., Zhang, X., Zhang, J., & Xie, L. (2018). Improved design of steel-concrete composite wind turbine tower under
uncertainty. Journal of Structural Engineering, 144, 8.
[59] Luo, H., Liu, L., Ye, K., Lu, Y., & Deng, X. (2020). Data-driven identification of critical factors affecting construction
labor productivity. Automation in Construction, 111, 2.
[60] Ma, L., Luo, X., Liu, X., & Wang, W. (2018). Optimal design of steel frame structures using artificial neural networks
and genetic algorithm. Journal of Construction Engineering and Management, 144, 4.
[61] Mendez, E., Masini, A., & Bez, H. E. (2019). Generative design methods for efficient building floor plan layouts.
Frontiers of Architectural Research, 8(2), 202–213.
[62] Mitchell, T. M. (1997). Machine Learning. McGraw-Hill.
[63] Nguyen, H., Kim, T., & Lee, J. (2018). Damage detection using deep learning from vibration data in a scaled bridge
model. Engineering Structures, 171, 698–707.
[64] Nguyen, T. H., Nguyen, V. T., & Nguyen, H. T. (2020). An AI-driven decision support system for structural health
monitoring of bridges. Engineering Structures, 204, 2.
[65] Nguyen, V. T., Tran, T. D., & Nguyen, D. T. (2018). Prediction of the ultimate strength of reinforced concrete beams
using artificial neural networks. Advances in Engineering Software, 117, 159–170.
[66] Parekh, Ruchit. Blueprint for Sustainability: LEED Implementation in Commercial Projects. Elsevier, 2024.
[67] Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning Representations by Backpropagating Errors.
Nature, 323(6088), 533–536.
[68] Parekh, Ruchit. Constructing Wellness: Harnessing AI for a Sustainable and Healthy Future. Elsevier, 2024.
[69] Parekh, Ruchit. "Trends and challenges in LEED v4. 1 healthcare certification: A comprehensive analysis of US
hospital scores in 2024." World Journal of Advanced Engineering Technology and Sciences 12.2 (2024): 726-740.
[70] Parekh, Ruchit, et al. "A Review of IoT-Enabled Smart Energy Hub Systems: Rising, Applications, Challenges, and
Future Prospects." (2024).

1079
 International Journal of Science and Research Archive, 2024, 13(01), 1059–1080

[71] Shan, Y., Guo, Q., Li, D., & Zhang, C. (2018). Optimization model for bike-sharing station configuration based on
evolutionary algorithm. Journal of Advanced Transportation, 20, 18.
[72] Shen, X., Huang, Z., Chen, W., Li, M., & Huang, Y. (2019). Truss optimization with a genetic algorithm based on the
finite element analysis. Symmetry, 11, 10.
[73] Smith, I. F., Elshafie, M. Z., & Sadullah, A. F. (2018). Machine learning applications in civil engineering. Civil
Engineering Journal, 4(5), 1130–1142.
[74] Sun, W., Chen, Q., Li, S., & Yu, R. (2019). Machine learning-based prediction of mechanical properties of cement-
based materials using X-ray micro-computed tomography images. Automation in Construction, 99, 34–45.
[75] Sun, Z., Huang, W., Zhou, Z., & Wang, J. (2020). Prediction model for cracking performance of asphalt pavements
using machine learning methods. Journal of Transportation Engineering, Part b: Pavements, 146, 3.
[76] Sutskever, I., Vinyals, O., & Le, Q. V. (2014). Sequence to sequence learning with neural networks. In Advances in
Neural Information Processing Systems (pp. 3104–3112).
[77] Szeliski, R. (2010). Computer Vision: Algorithms and Applications. Springer.
[78] Tabeshpour, M. R., Sheikhzadeh, M., Rahai, A. R., & Banan, M. R. (2021). Steel truss optimization using deep neural
network and genetic algorithm. Journal of Computing in Civil Engineering, 35, 1.
[79] Vapnik, V. (1995). The Nature of Statistical Learning Theory. Springer.
[80] Wang, D., & Liu, L. (2020). Construction project scheduling optimization using particle swarm optimization
integrated with cloud computing and machine learning. Automation in Construction, 110, 2.
[81] Wibowo, A., Roshani, S., Fosu, P. A., & Gospodini, A. (2020). Review of artificial intelligence in buildings:
Opportunities, challenges, and methodologies. Building and Environment, 182, 2.
[82] Wu, C., Yang, S., Xiong, H., & Ye, S. (2019). Artificial intelligence in civil engineering: Recent advances and future
trends. Frontiers of Engineering Management, 7(3), 332–342.
[83] Wu, S., Yi, Z., Li, Z., & Wang, Y. (2017). Application of machine learning approaches to material selection in
concrete pavement design. Journal of Computing in Civil Engineering, 31, 6.
[84] Wu, Z., Qian, H., & Li, L. (2016). Multi-objective optimization for steel truss design using a modified genetic
algorithm. Advances in Structural Engineering, 19(6), 1023–1034.
[85] Yin, K., Huang, W., & Xu, X. (2020). Landslide susceptibility analysis using deep neural networks: A case study in
Chongqing. China. Engineering Geology, 276, 2.
[86] Zahedi, A., Nafisi, A., & Afshar, A. (2020). Fuzzy analytic hierarchy process for material selection in bridge
construction. Journal of Performance of Constructed Facilities, 34, 2.
[87] Zhang, C., Wu, M., Zhou, K., Li, B., & Yan, G. (2020). Improved structure design optimization via neural network-
based surrogate modeling and parallel MOEA/D. Automation in Construction, 114, 2.
[88] Zhang, J., Gao, R. X., & Yan, R. (2015). An intelligent modal parameter identification method for structural health
monitoring. Structural Health Monitoring, 14(6), 618–633.
[89] Zhang, L., Li, S., Gong, Y., & Xie, Y. M. (2020). Shape optimization of truss structures using generative adversarial
network. Computer-Aided Design, 125, 2.
[90] Zhang, Y., Huang, C., Zhang, J., Chen, Y., & Wu, Z. (2020). Predicting the flexural behavior of FRP-reinforced
concrete beams using artificial neural networks. Advances in Civil Engineering, 20, 20.
[91] Zhao, X., Ren, W., & Li, Y. (2017). Fuzzy logic based computational intelligence for structural analysis. Structure
and Infrastructure Engineering, 13(2), 227–239.
[92] Zhu, J., Lan, T., & Wang, W. (2017). An AI-driven design exploration framework for optimization of pedestrian
bridge. Journal of Computing in Civil Engineering, 31, 5.

1080