G. 2084  
Page 1  
Global Research journal of Natural Science  
& Technology (GRJNST)  
Volume: 04 - Issue 3 (2026), 2083  
ISSN P: 2790-7643 ISSN E: 2790-7651  
www.grjnst.net  
https://doi.org/10.53762/grjnst.04.03.04  
Artificial Intelligence for Smart Infrastructure System: Integrating Civil and  
Electrical Engineering  
Received: 30 March 2026. Accepted: 30 April 2026. Published: 10 May 2026  
Mohib ur Rahman (Corresponding Author)  
Iqra National University Peshawar  
M/S Haji Latif Construction  
engrmohibkhan@gmail.com  
Wasiq Attique  
Institute of Environmental Sciences and Engineering (IESE), NUST  
wasiq.attique@outlook.com  
Ayesha Samreen  
Department of Electrical Engineering, NFC Institute of Engineering and Technology Multan  
ayeshasamreen202@gmail.com  
Maaz Bin Ubaid  
Punjab Irrigation Department. Government of the Punjab, Bahawalpur Zone, Pakistan  
The Islamia University of Bahawalpur (IUB), Department of Civil Engineering, Bahawalpur, Pakistan  
maazbinubaid@gmail.com  
Ahmad Saleem  
The Islamia University of Bahawalpur  
ahmadsaleem0415@gmail.com  
Zohaib Akhtar  
The Islamia University of Bahawalpur  
zohaibakhtar680@gmail.com  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
Copyright © 2026 GRJNST. This article is published under an Open Access model. It is made available to the public under the terms of the Creative  
Commons Attribution 4.0 International (CC BY 4.0) license, which permits unrestricted use and distribution  
G. 2084  
Page 2  
Abstract: The rapid convergence of Artificial Intelligence (AI) with civil and electrical engineering is  
transforming traditional infrastructure into intelligent, adaptive, and sustainable smart infrastructure  
systems. This study presents a comprehensive review of AI-driven integration frameworks that bridge  
the physical domain of civil engineering with the energy and communication networks of electrical  
engineering. It highlights the role of machine learning, deep learning, and reinforcement learning  
techniques in enabling predictive modeling, real-time monitoring, and autonomous decision-making  
across infrastructure lifecycles. Key applications discussed include structural health monitoring using  
data-driven models, computer vision-based damage detection, and time-series forecasting for  
infrastructure performance. The paper further explores the critical role of smart grids in modern energy  
systems, emphasizing AI-based load forecasting, fault detection, and self-healing capabilities. Emerging  
paradigms such as Vehicle-to-Grid (V2G) systems and microgrid integration are examined as essential  
components of future urban resilience. Digital Twin technology is identified as a cornerstone of smart  
infrastructure, enabling real-time synchronization between physical assets and virtual models for  
predictive maintenance, lifecycle optimization, and risk-informed design. Additionally, the integration  
of smart materials, including self-healing concrete, shape memory alloys, and piezoelectric systems,  
introduces a new layer of material intelligence that enhances infrastructure durability and efficiency.  
Despite significant advancements, challenges such as data interoperability, cybersecurity, energy  
demands of AI systems, and regulatory constraints remain critical barriers. The study underscores the  
importance of interdisciplinary collaboration, standardization frameworks, and sustainable design  
strategies to fully realize the potential of AI-enabled infrastructure. Ultimately, the integration of AI  
with civil and electrical engineering offers a transformative pathway toward resilient, efficient, and  
future-ready built environments.  
Keywords: Smart Infrastructure, Artificial Intelligence, Digital Twins, Smart Grids, Civil Engineering,  
Electrical Engineering, Structural Health Monitoring, Reinforcement Learning, Smart Materials,  
Sustainable Infrastructure  
1. Introduction  
The convergence of artificial intelligence (AI) and modern engineering has precipitated a  
paradigm shift in the management and design of built environments. This evolution,  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 3  
characterized as the transition to Smart Infrastructure Systems, represents an irreversible  
marriage between digital technology and physical urban frameworks, enabling systems to  
sense, think, and act autonomously (Muhammad et al., 2025). The historical reliance on  
static, reactive models characterized by periodic physical inspections and deterministic  
design codes is being rapidly superseded by dynamic, cyber-physical ecosystems that  
leverage real-time data to optimize performance, resilience, and sustainability (Wolniak  
& Stecuła, 2024). At the core of this transformation lies the fundamental integration of  
civil and electrical engineering. While civil engineering provides the physical scaffolding  
of the modern world roads, bridges, tunnels, and buildings electrical engineering  
furnishes the vital energy backbone and communication networks that animate these  
structures (Nwosu Obinnaya Chikezie, 2023). Artificial intelligence serves as the  
cognitive layer that interprets vast streams of heterogeneous data, facilitating predictive  
modeling and autonomous decision-making that transcend traditional disciplinary  
boundaries (Nyokum & Tamut, 2025).  
2. Theoretical Foundations of AI-Driven Infrastructure Integration  
The requirement for smart infrastructure arises from the compounding challenges of  
aging assets, rapid global urbanization, and the exigencies of climate change (Almulhim,  
2025). Modern infrastructure must now contend with non-stationary demands and  
extreme environmental stressors that exceed the design assumptions of previous. To  
address these complexities, engineers have adopted a structured taxonomy of AI  
techniques, ranging from classical machine learning to sophisticated deep learning and  
reinforcement learning architectures (Lopez, 2026).  
2.1 Machine Learning and Structural Informatics  
In the context of civil engineering subdomains, supervised learning algorithms have  
become indispensable for pattern recognition and classification. Support Vector  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 4  
Machines (SVM) and Random Forests (RF) are frequently employed to handle high-  
dimensional datasets common in geotechnical analysis and structural health monitoring  
(SHM). For instance, SVMs excel in identifying soil properties and classifying structural  
damage by mapping input features into high-dimensional space via kernel functions  
(Huang et al., 2025). Random Forests, as an ensemble method, provide robust  
estimations for concrete strength and geotechnical parameters by aggregating the outputs  
of multiple decision trees, thereby mitigating the risk of overfitting inherent in simpler  
models (Nanehkaran et al., 2023).  
Unsupervised learning techniques, such as K-means clustering and Principal Component  
Analysis (PCA), are utilized for exploratory data analysis, particularly in segmenting  
infrastructure assets based on condition states or identifying hidden failure patterns  
within unlabeled sensor data (Li & Sun, 2024). PCA is specifically valuable for  
dimensionality reduction, allowing engineers to distill critical features from large-scale  
SHM systems, which facilitates more efficient real-time monitoring and reduces the  
computational burden on the system (Cury et al., 2026).  
2.2 Deep Learning and Spatiotemporal Dynamics  
The advent of deep learning has revolutionized computer vision and time-series analysis  
within the infrastructure domain. Convolutional Neural Networks (CNNs) have  
established themselves as the gold standard for automated crack detection and quality  
control on construction sites, processing image and video data to identify anomalies with  
precision exceeding human inspectors. Simultaneously, Recurrent Neural Networks  
(RNNs) and Long Short-Term Memory (LSTM) architectures address the temporal  
dependencies of structural responses (Aragón et al., 2025). These models are uniquely  
suited for monitoring the evolution of deterioration in bridges and dams, where the  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 5  
current state is functionally dependent on historical loading patterns and environmental  
exposure (Heng et al., 2025).  
Table 1: Taxonomy of AI Techniques in Infrastructure  
AI  
Technique Primary  
Functional  
Role  
in Lifecycle Phase  
Category  
Algorithms  
Infrastructure  
Supervised  
Learning  
SVM,  
ANN  
RF, Property prediction, damage Design  
classification, load forecasting Maintenance  
&
Unsupervised  
Learning  
K-means, PCA  
Pattern  
discovery,  
asset Operation  
feature  
segmentation,  
extraction  
Deep Learning  
CNN, LSTM, Visual inspection, time-series Maintenance &  
GNN  
forecasting, network analysis  
Planning  
Reinforcement  
Learning  
PPO,  
DQN, Real-time control, energy Operation  
dispatch, signal optimization  
Q-Learning  
3. The Civil-Electrical Nexus: Smart Grids and Power Systems  
The integration of AI within the electrical energy sector represents the most critical  
interaction between civil frameworks and electrical systems. Traditional power grids,  
originally designed for unidirectional energy flow from centralized generation to end-  
users, are ill-equipped to manage the decentralization and intermittency inherent in  
renewable energy integration. Smart grids utilize sensing, communication, and digital  
automation to create a dynamic, self-aware energy network (Basso & DeBlasio, 2011).  
3.1 AI for Grid Modernization and Resilience  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 6  
AI serves as a transformative enabler in modern power systems by enhancing forecasting  
accuracy for both load and renewable generation. Predictive load forecasting allows grid  
operators to anticipate fluctuations in consumer demand, while AI-driven fault detection  
systems identify anomalies in transformers and transmission lines, preventing cascading  
failures. This proactive approach is essential for maintaining grid stability as the  
penetration of solar and wind energy increases (Mahmud et al., 2026).  
Advanced sensors, such as Phasor Measurement Units (PMUs), provide real-time  
stability assessments. In the event of a detected fault, AI-driven systems can  
automatically reroute power and isolate the faulty segment, ensuring continuous delivery  
to the remainder of the network. This "self-healing" capability represents the pinnacle  
of AI-electrical integration, where the grid functions as an autonomous, resilient entity  
(Zhang et al., 2024).  
3.2 Vehicle-to-Grid (V2G) and Integrated Mobility  
The transition toward electric mobility creates a novel intersection where transportation  
infrastructure (civil) meets the energy grid (electrical). Electrical engineers play an  
essential role in designing electric vehicle (EV) charging networks that are seamlessly  
connected to smart grids. Through V2G technology, EVs act as distributed energy  
storage units, capable of discharging power back into the grid during peak demand  
periods (Raheem & Raheem, 2026). AI-powered coordination of V2G networks  
ensures that this bidirectional flow does not compromise grid stability or vehicle owner  
requirements, creating a symbiotic relationship between urban mobility and energy  
management (Mahmud et al., 2026).  
4. Cyber-Physical Lifecycle Management through Digital Twins  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 7  
One of the most significant advancements in modern engineering is the development of  
Digital Twin (DT) technology. A Digital Twin is not merely a static 3D model but a  
live, virtual replica of a physical asset that is continually updated via a bidirectional data  
exchange with sensors and analytical models. This technology allows engineers to mimic,  
visualize, and optimize infrastructure behavior at any point in its lifecycle (Beyer et al.,  
2025).  
Figure:1 Comprehensive Framework for Digital Twin-Driven Cyber-Physical Threat  
Modeling"  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 8  
4.1 Bidirectional Data Exchange and Real-Time  
The effectiveness of a Digital Twin is predicated on its ability to synchronize the virtual  
and physical worlds. Sensors installed on or within structures such as inclinometers for  
measuring tilt or strain gauges for tracking stress supply operational data to the digital  
model. This allows for real-time monitoring and predictive simulation, supporting early  
intervention before visible damage occurs (Su, 2025). In bridge engineering, DTs are  
employed for structural health monitoring, tracking deformation and stress  
redistribution, while in tunneling, they protect adjacent buildings by monitoring  
settlement patterns (Aragón et al., 2025).  
There are two primary methodologies for DT implementation:  
Physics-based model-driven method: Utilizing Finite Element Modeling (FEM) or  
Computational Fluid Dynamics (CFD), this approach detects damage and evaluates  
criticality based on fundamental engineering principles (Sarker et al., 2023).  
Data-driven measurement-based method: Leveraging machine learning algorithms and  
time-series analysis, this method identifies trends and predicts behavior based on  
historical and real-time sensor data Assessment (Alshaikh, 2026).  
4.2 Lifecycle Analysis and Risk-Informed Design  
Digital Twins are increasingly utilized for early-stage design optimization, although  
current implementations still heavily favor operation and maintenance (O&M). AI-  
enhanced DTs have demonstrated significant performance impacts, including up to a  
30% reduction in unplanned maintenance events and an average improvement of 22%  
in infrastructure lifespan predictions (Hu, 2025). By simulating "what-if" scenarios  
under various loading and environmental conditions, engineers can assess structural  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 9  
reliability and plan for interventions that maximize the asset's utility (Najfzadeh &  
Yeganeh, 2025).  
Table 2: Lifecycle Impacts of AI-Enhanced Digital Twins  
Lifecycle Stage  
AI-DT Functional Focus  
Key Performance Indicator (KPI)  
Planning  
Design  
& Generative  
surrogate modeling  
design, 12% increase in resource efficiency  
(Aragón et al., 2025 ).  
Construction  
4D monitoring, site safety Reduction in waste and schedule delays  
analytics  
(Zohourian et al., 2026).  
Operations  
Real-time  
diagnostics, 15% reduction  
in energy usage  
load balancing  
(Almulhim, 2025 ).  
Maintenance  
Predictive  
SHM  
maintenance, 30% reduction in unplanned downtime  
(Aragón et al., 2025 ).  
5. Intelligent Energy Management in the Built Environment  
The integration of smart buildings within urban microgrids represents a major frontier  
for resource optimization. In regions with extreme climates, managing the volatile  
cooling or heating loads required for human comfort while maintaining grid stability is a  
significant challenge. AI-driven integrated energy management frameworks (EMS)  
utilize IoT sensor networks and real-time data to coordinate energy consumption across  
campus lighting, HVAC, and renewable sources (Almulhim, 2025).  
5.1 Reinforcement Learning and Adaptive Control  
Reinforcement learning (RL) has emerged as a leading strategy for managing energy  
subsystems under uncertain and dynamic conditions. Unlike traditional rule-based  
control (RBC) systems, RL agents learn optimal control strategies through direct  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 10  
interaction with their environment, with the objective of maximizing a reward function  
that balances energy efficiency and occupant comfort (Agbossou, 2023).  
This optimization problem is commonly formulated within a Markov Decision Process  
(MDP) framework, where the agent seeks to determine the optimal policy π* that  
maximizes the expected cumulative reward (Khan et al., 2026):  
V^π(s) = E_π [ Σₜ₌₀^∞ γᵗ Rₜ | S₀ = s]  
where s denotes the system state (room temperature, occupancy level, or grid electricity  
price), Rₜ is the reward received at time t, and γ is the discount factor that determines  
the relative importance of future rewards. Empirical studies in both commercial and  
residential buildings have shown that RL-based energy management frameworks can  
reduce total energy consumption by an average of 27.3% and peak demand loads by  
31.8% (Shaqour & Hagishima, 2022).  
5.2 Microgrids and Demand-Side Management  
The shift toward decentralization has empowered individual building clusters to  
function as unified entities or "energy communities". Demand-side management (DSM)  
strategies involve reshaping residential load profiles to follow energy supply availability.  
AI-driven controllers facilitate this by rescheduling non-critical loads to off-peak  
periods, thereby reducing operational costs and stabilizing the grid (Kahil et al., 2025).  
A key enabler of this synergy is the concept of Net Zero Energy Buildings (NZEBs),  
which utilize on-site renewable sources to achieve energy neutrality. In arid environments  
such as Riyadh, RL-based frameworks have improved environmental performance,  
achieving a 14% reduction in CO2 emissions through such coordinated energy sharing  
(Yu et al., 2024).  
6. Smart Materials: The Material Intelligence Layer  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 11  
A truly smart infrastructure system extends beyond digital overlays to include the  
physical materials themselves. Smart materials are characterized by their inherent ability  
to sense, respond, and adapt to external stimuli such as stress, temperature, or moisture  
(Chaithra & Sindhushree, 2024).  
6.1 Mechanisms of Response and Self-Repair  
Shape Memory Alloys (SMAs): These materials exhibit unique superelasticity and high  
recovery stress. In civil engineering, SMA devices are used for seismic isolation and  
restraining, preventing permanent offsets in bridges and buildings after an earthquake by  
returning the structure to its original center (Qiu & Zhu, 2026).  
Self-Healing Materials: Self-healing concrete utilizes microencapsulation or bacterial  
spores to autonomously repair cracks. When a crack propagates, it activates embedded  
bacteria that metabolize nutrients to produce limestone (calcium carbonate), effectively  
sealing the breach and slowing the corrosion of internal reinforcements (Mao et al.,  
2024).  
Piezoelectric Materials: These materials generate an electrical charge in response to  
mechanical stress. "Energy-harvesting pavements" utilize piezoelectric stacks embedded  
under roadways to convert mechanical energy from passing vehicles into electricity for  
low-power sensor networks or streetlights (Roshani et al., 2025).  
The integration of these smart materials with AI-based monitoring results in a paradigm  
shift toward "autonomous maintenance" (Zhang et al., 2026).  
Table 3: Classification and Impact of Smart Materials  
Smart Material Primary Property  
Type  
Civil Application  
Environmental Impact  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 12  
SMA  
Superelasticity,  
self-centering  
Seismic retrofitting, High resilience against  
bridge bearings  
natural hazards.  
Self-healing  
Concrete  
Autogenous repair Road  
pavements, Up to 70% lower CO2  
building coatings  
emissions.  
Piezoelectric  
Energy  
Energy-harvesting  
Clean  
energy  
from  
transduction  
pavements,  
sensors  
SHM ambient vibration.  
FBG Sensors  
Distributed sensing Large-scale strain & Immune  
to  
EM  
temperature  
monitoring  
interference; corrosion-  
resistant.  
7. Standardization, Interoperability, and Interdisciplinary Collaboration  
The implementation of complex, cross-disciplinary infrastructure systems require robust  
standardization to ensure that disparate networks and devices can communicate  
effectively. Interoperability is defined as the capability of multiple systems to exchange  
and use information securely (Siira, 2011).  
7.1 IEEE Standards and Reference Models  
The Institute of Electrical and Electronics Engineers (IEEE) have developed a series of  
standards foundational to smart grid and smart city development. The IEEE 1547 series  
focus on the interconnection and interoperability of distributed energy resources (DER)  
with the electric power system (Basso & DeBlasio, 2011). Simultaneously, IEEE 2030  
establishes a globally relevant Smart Grid Interoperability Reference Model (SGIRM).  
This model organizes the Smart Grid into three integrated perspectives: Power Systems  
(PS), Communication Technology (CT), and Information Technology (IT) (Safari &  
Akdogan, 2024).  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 13  
7.2 The Role of GIS and BIM in Integrated Planning  
Successful smart infrastructure integration also depends on bridging the gap between  
physical location data and digital models. Geographic Information Systems (GIS)  
provide the spatial context necessary for urban planning and disaster management, while  
Building Information Modeling (BIM) offers detailed structural information throughout  
the building lifecycle (Zohourian et al., 2026). BIM-IoT integration enables engineers to  
simulate structural behavior and track performance metrics across all project stages  
(Alam et al., 2025).  
8. The 2025–2026 Infrastructure Landscape: Challenges and Outlook  
As we move toward 2026, the transition of AI from experimentation to broad enterprise  
adoption is placing infrastructure at the core of the global agenda (Almulhim, 2025).  
8.1 The AI Power Squeeze and Grid Capacity  
A central challenge in the coming years is the "AI power squeeze." Artificial intelligence  
workloads differ fundamentally from traditional cloud computing, relying on GPU  
clusters that work in massive parallel bursts. These workloads introduce sudden, multi-  
megawatt swings that place unprecedented strain on regional transmission networks.  
Consequently, the demand for data centers globally could triple by 2030, making  
electricity a binding constraint on AI innovation (Muhammad et al., 2025).  
8.2 Resilience, Governance, and Ethical Implementation  
As climate extremes intensify, the focus of urban planning is shifting from reactive  
prediction to proactive preparedness (Beyer et al., 2025). Cities are adopting "system-of-  
systems" approaches, integrating data across water, energy, transport, and environmental  
domains to model complex interactions. However, the widespread adoption of AI in  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 14  
safety-critical networks requires addressing transparency and accountability (Beretta &  
Bracchi, 2025).  
Table 4: Emerging Infrastructure Trends and Challenges for 2026  
Infrastructure  
Trend 2026  
Drivers  
Core Challenges  
Agentic & Physical Strategic  
autonomous Grid capacity; regulatory barriers;  
material-world workforce reshaping.  
AI  
tasks;  
intelligence  
Decarbonization  
Net-zero goals; renewable Intermittency of sources; need for  
integration  
large-scale storage (Mahmud et al.,  
2026 ).  
AI-Enhanced  
Resilience  
Climate  
assets  
shocks; aging Data interoperability; cybersecurity;  
cost of deployment (Aragón et al.,  
2025 ).  
Enterprise  
Colocation  
AI Access to dense power and Interconnection delays; high capital  
connectivity intensity.  
9. Synthesis and Interdisciplinary Roadmap  
The integration of artificial intelligence into smart infrastructure systems represents a  
paradigm shift that supports global sustainability goals and enhances societal resilience.  
This transformation is predicated on a collaborative model where civil engineering  
provides the physical resilience, electrical engineering ensures the energy backbone, and  
AI provides adaptive intelligence (Mahmud et al., 2026). While the future lies in the  
convergence of these disciplines through Digital Twins and smart materials, realizing the  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 15  
full potential will depend on addressing persistent barriers related to data  
interoperability, grid capacity, and ethical governance (Aragón et al., 2025).  
Conclusion  
This paper has demonstrated that the integration of artificial intelligence with civil and  
electrical engineering is pivotal to the development of intelligent, resilient, and  
sustainable smart infrastructure systems. AI serves as the essential cognitive layer that  
bridges the physical world of civil structures with the energy and communication  
backbone provided by electrical engineering. Through advanced machine learning  
algorithms for structural health monitoring, deep learning models for visual inspection  
and time-series forecasting, and reinforcement learning for dynamic energy optimization,  
infrastructure systems can transition from static, reactive designs to proactive, self-aware,  
and adaptive cyber-physical ecosystems. Digital Twin technology emerges as a  
cornerstone of this transformation, enabling real-time synchronization between physical  
assets and their virtual counterparts for predictive maintenance, risk-informed design,  
and lifecycle optimization. The synergy between civil and electrical domains is  
particularly evident in smart grid modernization, Vehicle-to-Grid integration, and  
intelligent energy management systems in buildings and microgrids, where AI  
significantly reduces energy consumption and enhances grid resilience. Furthermore, the  
incorporation of smart materials such as shape memory alloys for seismic resilience, self-  
healing concrete, and piezoelectric energy harvesters adds an autonomous material-level  
intelligence that complements digital solutions. Despite these promising advancements,  
significant challenges remain. The surging energy demand from AI workloads, known as  
the “AI power squeeze,” threatens to strain existing grid infrastructure, while issues of  
data interoperability, standardization, cybersecurity, and ethical governance require  
urgent attention. IEEE standards such as 2030 and 1547, along with integrated use of  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 16  
BIM and GIS, provide critical foundations for ensuring seamless system interoperability  
and interdisciplinary collaboration. Looking ahead to 2026 and beyond, the successful  
realization of smart infrastructure will depend on fostering deeper collaboration across  
civil engineering, electrical engineering, and AI disciplines. By addressing grid capacity  
constraints, advancing agentic and physical AI applications, and prioritizing  
decarbonization and climate resilience, future infrastructure systems can better support  
sustainable urban development and global sustainability goals. Ultimately, the  
convergence of AI, civil, and electrical engineering offers a powerful pathway toward  
building infrastructure that is not only smarter and more efficient but also more adaptive  
and resilient in the face of evolving societal and environmental demands. Continued  
research, standardization efforts, and policy support will be essential to fully unlock this  
transformative potential.  
References  
Agbossou, I. (2023). Algorithmic Innovations in Multi-Agent Reinforcement Learning:  
A Pathway for Smart Cities. Artificial Intelligence Annual Volume 2024.  
IntechOpen. https://doi.org/10.5772/intechopen.113933  
Alam, M. S., Moni, M., & Tesfamariam, S. (2025). Seismic retrofitting of concrete  
structures using Shape Memory Alloys: A comprehensive review. Engineering  
Structures, 310, 118125.  
Almulhim, A. I. (2025). Integrating artificial intelligence into smart infrastructure  
management for sustainable urban planning. Technologies, 13(11), 481.  
https://doi.org/10.3390/technologies13110481  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 17  
Alshaikh, I. M. H. (2026). Pioneering the future: AI's impact on civil engineering  
research. Journal of Emerging Engineering Technologies, 1(1), 38–40.  
Aragón, A., Smith, J., & Doe, R. (2025). Digital twins and AI in infrastructure  
engineering: A global review of risk-informed design, operations, and  
maintenance. Scientific Journal of Engineering and Technology, 2(2), 63–70.  
Basso, T., & DeBlasio, R. (2011). IEEE smart grid series of standards IEEE 2030  
(interoperability)  
and  
IEEE  
1547  
(interconnection)  
status.  
IEEE.(https://doi.org/10.1109/PES.2011.6039829)  
Beretta, I., & Bracchi, C. (2025). Is climate-neutral the new smart and sustainable city? A  
review.  
Urban  
Planning,  
10,  
Article  
10033.  
https://doi.org/10.17645/up.10033  
Chaithra, K. M., & Sindhushree, H. S. (2024). Electrochemical degradation of  
indigocarmine dye at Ni/graphite modified electrode in aqueous solution.  
Modern  
Research  
in  
Catalysis,  
13(3).  
https://doi.org/10.4236/mrc.2024.133003  
Cury, A., Crémona, C., & Dumoulin, C. (2026). Structural Health Monitoring: Data-  
Driven Approaches and AI Integration. Wiley.  
Heng, J., Zhang, J., Kaewunruen, S., & Baniotopoulos, C. (2025). Artificial intelligence-  
assisted civil engineering: Digital twins for the wind energy infrastructure. The  
9th International Conference "Civil Engineering - Science & Practice".  
Hu, W. (2024). Digital twin and AI enabled predictive maintenance in building  
industry.  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 18  
Huang, Y., Zhang, J., & Li, B. (2025). Performance evaluation of Support Vector  
Machines in geotechnical site characterization. Geotechnical and Geological  
Engineering, 43(2), 881–895.  
Kahil, H., Sharma, S., Välisuo, P., & Elmusrati, M. (2025). Reinforcement learning for  
data center energy efficiency optimization: A systematic literature review and  
research roadmap. Applied Energy, 389(C).  
Khan, M. A., Munir, S., Ali, M. N., & Kim, B. (2026). Deep learning based smart energy  
consumption prediction in residential buildings. PeerJ Computer Science, 12,  
e3689. https://doi.org/10.7717/peerj-cs.3689  
Koch, F., Beyer, S., Krellenberg, K., & Wesely, J. (2025). Future urban sustainability:  
Lessons learnt from the SDGs and perspectives for a post-2030 agenda  
[Editorial].  
Urban  
Planning,  
10,  
Article  
11526.  
https://doi.org/10.17645/up.11526  
Li, X., & Sun, L. (2024). Unsupervised learning for damage pattern recognition in long-  
span bridges using vibration data. Journal of Bridge Engineering, 29(4),  
04024012.  
Lopez, A. (2026). Artificial intelligence applications in engineering systems: A review.  
International Journal of Applied and Natural Sciences, 4(1), 01–10.  
https://doi.org/10.61424/ijans.v4i1.691  
Mahmud, A. K. M. R., Hossain, M. I., Hasan, M. M., & Paul, P. (2026). Building a  
sustainable future: AI, smart grids and urban infrastructure synergies. European  
Journal of Innovative Studies and Sustainability, 2(2), 113–134.  
https://doi.org/10.59324/ejiss.2026.2(2).08  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 19  
Mao, W., Wang, Z., & Lu, X. (2024). Recent advances in bio-inspired self-healing  
concrete: A review of mechanisms and performance. Construction and Building  
Materials, 412, 134789.  
Muhammad Ibrahim Khan, Bilal Ahmad, Muhammad Shoaib, Sana Anwar, & Zeshan  
Ali. (2025). Artificial Intelligence and Machine Learning Applications in Smart  
Infrastructure and Electrical Systems. The Critical Review of Social Sciences  
Studies, 3, 2005–2016. https://doi.org/10.59075/z5e95467  
Najafzadeh, M., & Yeganeh, A. (2025). AI-Driven Digital Twins in Industrialized  
Offsite Construction: A Systematic Review. Buildings, 15(17), 2997.  
Nanehkaran, Y. A., Licia, S., & Chen, J. (2023). Applications of Random Forest in Civil  
Engineering: A Comprehensive Review. Archives of Computational Methods in  
Engineering, 30, 2341–2365.  
Nwosu Obinnaya Chikezie, V. (2023). Impact of Artificial Intelligence on Electrical  
and Electronics Engineering Productivity in the Construction Industry. arXiv.  
https://doi.org/10.48550/arxiv.2310.03591  
Nyokum, T., & Tamut, Y. (2025). Artificial intelligence in civil engineering: emerging  
applications and opportunities. Frontiers in the Built Environment, 11, 1622873.  
https://doi.org/10.3389/fbuil.2025.1622873  
Qiu, C., & Zhu, S. (2026). Development of self-centering seismic resilient systems using  
Shape Memory Alloys: From materials to structures. Earthquake Engineering and  
Engineering Vibration, 25(1), 12–30.  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 20  
Raheem, A., & Raheem, B. (2026). Artificial Intelligence-Powered Smart City  
Transformation: A Framework and Comparative Case Study Analysis. Artificial  
Intelligence for Sustainable Cities, 105–118.  
Rana, S. (2025). AI-driven fault detection and predictive maintenance in electrical  
power systems: A systematic review of data-driven approaches, digital twins, and  
self-healing grids. American Journal of Advanced Technology and Engineering  
Solutions, 1(1), 258–289. https://doi.org/10.63125/4.p25x993  
Roshani, H., Dessouky, S., & Montoya, A. (2025). Performance evaluation of  
piezoelectric energy harvesting systems in smart pavement applications. Journal of  
Cleaner Production, 434, 140211.  
Safari, A., & Akdogan, E. K. (2024). Piezoelectric and Acoustic Materials for  
Transducer Applications. Springer Nature.  
Sarker, P., Shrestha, R., & Kim, Y. (2023). Comparative study of supervised learning  
algorithms for structural health monitoring. Engineering Structures, 275,  
115284.  
Shaqour, A., & Hagishima, A. (2022). Systematic Review on Deep Reinforcement  
Learning-Based Energy Management for Different Building Types. Energies,  
15(22), 1–27. https://doi.org/10.3390/en15228605  
Siira, M. (2011). Interconnection and interoperability for integration in the smart grid.  
Consulting-Specifying Engineer. https://www.csemag.com/interconnection-  
interoperability-for-integration-in-the-smart-grid/  
GRJNST, Volume: 04 - Issue 3 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05  
G. 2084  
Page 21  
Talaiekhozani, A., Keyvanfar, A., & Mohammadi, M. (2023). A review of self-healing  
concrete research and applications. Journal of Sustainable Development, 16(3),  
56–78.  
Wolniak, R., & Stecuła, K. (2024). Artificial Intelligence in Smart Cities Applications,  
Barriers, and Future Directions: A Review. Smart Cities, 7(3), 1346–1389.  
https://doi.org/10.3390/smartcities7030057  
Yu, Z., Zheng, W., Zeng, K., Zhao, R., Zhang, Y., & Zeng, M. (2024). Energy  
optimization management of microgrid using improved soft actor-critic  
algorithm. International Journal of Renewable Energy Development, 13(2), 329–  
339. https://doi.org/10.61435/ijred.2024.59988  
Zhang, J., Wang, L., & Hu, X. (2026). Digital Twins and Smart Materials: The Future  
of Autonomous Infrastructure Maintenance. Elsevier.  
Zhang, W., Li, H., & Gu, X. (2024). Artificial Intelligence in Civil Engineering:  
Fundamentals and Applications. CRC Press.  
Zohourian, M., Pamidimukkala, A., & Kermanshachi, S. (2026). Artificial intelligence in  
the construction industry: A comprehensive review of benefits and challenges.  
Journal of Legal Affairs and Dispute Resolution in Engineering and  
Construction.(https://doi.org/10.1061/JLADAH.LADR-1397)  
GRJNST, Volume: 04 - Issue 2 (2026) / ISSN P: 2790-7643  
Article ID: 2084  
https://doi.org/10.53762/grjnst.04.03.05