Constructing a New Future Through Artificial Intelligence

Although still in its infancy, artificial intelligence (AI) is already transforming built environments all around the globe. Integrated into such technologies as sensors, drones and advanced new construction materials, AI enables the design, construction, operation and maintenance of smarter, safer and more sustainable large-scale infrastructure than previously possible.

 Dedicated to developing cross-disciplinary solutions to infrastructure challenges, Stevens Institute of Technology’s Department of Civil, Environmental and Ocean Engineering sits at the forefront of this revolutionary new future.

From designing greener, longer-lasting concrete to monitoring bridges via adaptive sensing systems, the department’s researchers are harnessing AI to develop tools for analyzing complex data, predicting risk and guiding critical decision-making. 

Here we highlight just a few of the department’s efforts to drive the next generation of intelligent infrastructure systems through AI in construction.

A holistic, self-adaptive system for damage monitoring and inspection

[Yi Bao/Kaijian Liu photos]
Research focus: Advanced sensing technologies, robotics and artificial intelligence for real-time monitoring and inspection of civil infrastructure

Research focus: Artificial intelligence technologies to inform data-driven optimization of the operation and maintenance phases of the construction lifecycle

With funding from national agencies including the Department of Transportation and Department of Agriculture, Yi Bao’s research focuses on developing advanced technologies — sensors, drones and artificial intelligence — to monitor and inspect large-scale infrastructure such as bridges, buildings and pipelines. Kaijian Liu — with projects sponsored by the United States Geological Survey, National Oceanic and Atmospheric Administration, and U.S. Army, amongst others — develops state-of-the-art sensing, artificial intelligence and cyber-physical systems for resilient and sustainable civil infrastructure. 

One of Liu and Bao’s flagship projects explores an AI-powered “mutualistic” cyber-physical system for self-adaptive multi-damage inspection and monitoring of civil infrastructure. Spearheaded by Liu as lead principal investigator, the project is funded by the National Science Foundation.

This system will focus on enabling a single type of sensor to reconfigure itself to monitor and detect multiple kinds of infrastructure damage. Drawing inspiration from mutualism in biology — where two species interact in a way that benefits both — the cyber side (AI-based damage prediction) and the physical components (reconfigurable sensors) will together enhance the overall system’s ability to predict the onset and propagation of damage, adapt according to need, and improve accuracy and performance.

Distributed fiber optic sensors are used to collect real-time data on infrastructure strain, temperature and vibration, while drones provide remote visual inspections in areas without sensors. Machine learning algorithms detect early signs of structural anomalies, classify types of damage and predict future deterioration. Results are fed into a digital twin model that analyzes infrastructure behavior in real time, identifying subtle patterns and making predictive assessments about structural health to enable informed decision-making for optimal operation and maintenance management. The system will be tested using both simulations and physical prototypes.

By reducing the numbers and types of sensors required to monitor multiple types of damage through mutualistic integration of cyber AI prediction and physical reconfigurable sensing, the system will reduce costs while increasing the timeliness and accuracy of damage assessment and prediction at the same time.

 “I imagine AI can help us build facilities in outer space, where humans can barely operate and construct. In video games, players can easily construct buildings, infrastructure and other structures by simply clicking a few buttons. Imagine designers, engineers and builders sitting somewhere on earth, sending intents and instructions to robots who can accurately understand these instructions and translate them into actions, eventually fully constructing buildings on the moon. Is this too bold? Maybe or maybe not. Let’s see.” – Kaijian Liu

This more cost-effective approach, said Liu, promises to “leapfrog the development of safer, more resilient and sustainable infrastructure, which would stimulate economic growth and social welfare for the benefit of the nation and its people.”

AI plays an indispensable role in this system, detecting patterns and processing and interpreting enormous amounts of high-dimensional data from a wide variety of sources in real time at a rate and accuracy that would be impossible to achieve manually.

The project, Bao said, has deepened his appreciation of AI’s potential in the construction and infrastructure domain.

“Initially, I viewed AI primarily as a computational tool,” he said. “Now, I see it as a transformative enabler that can shift our entire approach to how we monitor, inspect and maintain infrastructure — from reactive to proactive, and from labor-intensive to data-driven.”

“In the near future, we will see self-monitoring infrastructure systems capable of assessing their own health, recommending maintenance and even self-adapting to changing conditions,” said Bao. “On construction sites, AI will support autonomous construction equipment, real-time quality control and intelligent resource allocation — greatly improving efficiency and safety. Combined with digital twins, AI will allow us to simulate and optimize infrastructure behavior across its entire lifecycle, enabling sustainable and resilient designs.”

Smarter building materials made better

[Weina headshot]

Research focus: Sustainable and resilient construction materials using AI-powered design frameworks

Weina Meng studies how to make concrete and other building materials stronger, greener and longer-lasting, using artificial intelligence to guide, accelerate and streamline the process. One such project focuses on incorporating various solid waste materials — including waste glass, recycled concrete and industrial by-products — to develop ultra-high-performance concrete (UHPC) that reduces carbon emissions and performs optimally under even the most aggressive environmental conditions.

Leveraging extensive datasets on material physical properties, chemical compositions and durability indicators, Meng applies data-driven methods to predict the performance of the various waste-derived constituents in custom-designed UHPC mixtures. She is also exploring combining AI with blockchain to track the origins, quality and carbon footprints of construction materials to provide greater transparency and accountability in the construction supply chain. Her research holds significant implications for durable and sustainable construction in building, transportation and coastal infrastructure applications.

“AI is essential in this work because the number of possible mix designs and material combinations is nearly infinite,” Meng said. “Traditional, experiment-based trial-and-error methods would take years, but AI models can rapidly identify optimal combinations by learning patterns and making accurate predictions. It has allowed us to accelerate discovery, improve performance and reduce material waste significantly.”

 “Looking ahead, I believe AI will transform construction by enabling real-time decision-making, autonomous quality control and smarter use of waste materials,” said Meng. “We’ll see personalized, adaptive infrastructure that is optimized for performance, cost and environmental impact. AI could also help us close the loop on construction waste, making circular economy practices the norm.”

Thanks to her recent research, Meng’s view of AI has evolved, she said, from a “powerful but somewhat opaque” pattern-recognition tool to one that, “when informed by domain knowledge, can play a transformative role in advancing both innovation and sustainability in construction materials.”

“By integrating fundamental principles of materials science into the AI framework, we’ve developed models that not only predict performance outcomes, but also offer interpretable insights aligned with known physical and chemical mechanisms,” she said. “This has shown me that AI doesn’t need to function as a black box. It can be guided, structured and aligned with engineering knowledge to enhance reliability and decision-making in construction applications.”

Integrating the classroom with the construction site

[Mohammad Ilbeigi headshot]

Research focus: Technology-enhanced, AI-informed methods for the sustainable and resilient development of urban infrastructure systems

Currently leading multiple projects funded by the National Science Foundation (NSF), Mohammad Ilbeigi’s research focuses on harnessing AI technologies to build smarter cities capable of withstanding widespread challenges. He also has an interest in developing learning methods that supplement traditional teachings beyond the classroom.

 “AI is rapidly transforming job qualifications in construction and may significantly disrupt traditional employment structures,” he noted. “Advancements in AI demand a fundamental transformation in workforce development across the construction industry. Civil engineering education must integrate AI literacy, data science and automation into core curricula to prepare future engineers for AI-driven workflows." – Mohammed Ilbeigi

One such project, funded by NSF’s Research on Innovative Technologies for Enhanced Learning program, employs AI-based system development to turn spontaneous everyday encounters into extracurricular learning opportunities.

Initially developed as a mobile app for civil engineering students, this fully automated system called CELENS (Civil Engineering Lens) will analyze high-dimensional visual and audio data from construction sites to identify and explain construction operations, equipment, structural elements and materials encountered in real time.

“For instance, if a student encounters a construction site while walking to class,” he explained, “they can use the app, and CELENS will explain what they are seeing, which enables experiential learning outside traditional classroom settings.”

Designed to run on basic smartphones and leverage a phone’s built-in microphone, the app will also link the students’ observations to their course materials and provide feedback about those observations to their instructors. The system will be implemented and tested in multiple undergraduate construction engineering courses at Stevens.

“The AI component of the CELENS platform enables on-demand access to educators independent of time, location or reliance on other resources,” Ilbeigi said. “It introduces a new way of learning from daily observations — an approach that would not be feasible without an intelligent AI system.” 

 Future phases will expand CELENS for use by construction managers and other professional decision-makers to generate a variety of reports, including progress updates and safety monitoring summaries. The app will be easily expandable to other industries and fields of study.

 “The CELENS platform makes the entire process fully automated, objective and trustworthy — capabilities that are not achievable without AI,” Ilbeigi said. “The more you engage with advancements in AI across various applications, the more evident it becomes how rapidly things are evolving and how many new opportunities are emerging for more intelligent and successful construction operations and management,” he said.

The future of AI in construction

 All four researchers agreed that AI is destined to transform the future of construction engineering and the ways that built environments are designed, constructed, managed and maintained.

 As for where AI will take us next in construction, the researchers highlighted a variety of likely possibilities.

 “Looking ahead, I believe AI will transform construction by enabling real-time decision-making, autonomous quality control and smarter use of waste materials,” said Meng. “We’ll see personalized, adaptive infrastructure that is optimized for performance, cost and environmental impact. AI could also help us close the loop on construction waste, making circular economy practices the norm.”

 “In the near future, we will see self-monitoring infrastructure systems capable of assessing their own health, recommending maintenance and even self-adapting to changing conditions,” said Bao. “On construction sites, AI will support autonomous construction equipment, real-time quality control and intelligent resource allocation — greatly improving efficiency and safety. Combined with digital twins, AI will allow us to simulate and optimize infrastructure behavior across its entire lifecycle, enabling sustainable and resilient designs.”

 Liu, meanwhile, looked to the skies for an ambitious vision of where AI could take us next.

 “I imagine AI can help us build facilities in outer space, where humans can barely operate and construct. In video games, players can easily construct buildings, infrastructure and other structures by simply clicking a few buttons. Imagine designers, engineers and builders sitting somewhere on earth, sending intents and instructions to robots who can accurately understand these instructions and translate them into actions, eventually fully constructing buildings on the moon. Is this too bold? Maybe or maybe not. Let’s see.”

 Ilbeigi agreed that the role and benefits of AI in future innovations are beyond question. AI, he said, is on track to “reshape roles, workflows and collaboration across the industry, making construction safer, faster and more resilient.”

 However, he also highlighted the importance of prioritizing how AI is integrated more broadly – into society, education and the workforce at large.

 “AI is rapidly transforming job qualifications in construction and may significantly disrupt traditional employment structures,” he noted. “Advancements in AI demand a fundamental transformation in workforce development across the construction industry. Civil engineering education must integrate AI literacy, data science and automation into core curricula to prepare future engineers for AI-driven workflows. At the same time, reskilling programs for construction workers should focus on digital tools, robotics and human–AI collaboration to ensure they can adapt to evolving site operations. Empowering both professionals and tradespeople with AI competencies is essential to fully realize the benefits of intelligent construction and maintain a competitive, future-ready workforce.”