Built smarter – AI and the future of infrastructure podcast

Transcript

Built smarter – AI and the future of infrastructure

Rebecca Torr (RT): Hi, I'm Rebecca Torr, and welcome to the Emerald Podcast Series. Today, we're exploring how artificial intelligence is transforming the built environment, and what that means to creating a more sustainable, resilient future. From smart cities and digital twins to AI driven construction and infrastructure management. We're seeing rapid innovation in civil engineering, but how can these technologies be used in ways that make a real impact? To help us unpack this, I'm joined by Dr Yelda Turkan, Associate Professor at Oregon State University and an expert in AI applications, in construction and infrastructure systems. She's also the co-editor in chief of Artificial Intelligence for a Sustainable Built Environment, a new open access journal from Emerald. In this episode, we'll discuss the most promising uses of AI in engineering today, the social and environmental implications, and the kinds of interdisciplinary research needed to shape a better future. Let's get into it.

Yelda Turkan (YT): What excites me most is that AI is shifting us from being more reactive to being more proactive in how we design and manage the built environment. So, instead of waiting for things to break or resources to be wasted, AI enables us to actually predict issues before they happen and optimise how we use materials and energy and create systems that adapt in real time. For example, imagine energy grids that balance themselves dynamically, or buildings that actively learn how to reduce their own environmental footprint, that move from static infrastructure to living adaptive systems is what I find transformative.

RT: I mean, it sounds so complex, and it is so complex, and people like yourself with the expertise, you know, it's incredible what you can actually imagine and what you can achieve. And thank you for speaking to us in sort of just really plain terms, because it is such a technical subject, but I think it building on what you were saying. I mean, there's a huge potential here. And you know, we have had an increasing amount of challenges in the built environment, and it's only going to grow with climate change and just all the constraints and the natural disasters and the frequency of events that we're having. So from your perspective, what are some of the challenges and the issues today that AI and smart technologies can help address?

YT: I think one of the biggest challenges that we face today is sustainability. Our buildings and infrastructure account for a huge amount of global carbon emissions, and AI can actually help us optimise energy use, reduce material waste, and also extend the life of our assets, infrastructure. Another challenge is resilience. As you mentioned, climate change is creating new risks, and AI can help us simulate scenarios, assess vulnerabilities and strengthen our responses to those challenges. And obviously there is the productivity challenge. As we all know, construction is still one of the least digitised sectors, and hopefully AI can help us streamline workflows and improve safety while reducing delays and cost overruns.

RT: I think from what you've just said, you know, obviously construction, it's notoriously associated with sort of like slow progress and, you know, lack of change and not necessarily having the know-how, or the resource to even auto change, you know, systems obviously can cost a lot of money. So it'd be interesting to sort of understand, you know, sort of what role AI can play in that. I mean, the fact that, you know, I don't know how expensive that would be, you know, those sort of AI systems, and, you know, to what extent let their the development of construction, you know, what extent they're sort of developed, and sort of, if they're ready made to just be introduced, or how easy it is to introduce into construction or into industry, specifically, you know, just to be taken and used. I don't know if you've got any sort of insights into that, and sort of how AI is being used in industry, and if it's starting to be really used, you know, sort of on the ground?

YT: I think it's being used in silos, for example, to optimise schedules or for tasks such as converting LiDAR point class into three dimensional models. It's being used in silos, but there's still, I think there are many challenges. We need more label data sets, more case studies that use. I think overall, we just need more data sets or develop algorithms for specific use cases. And also, there is the transparency issue that the or the trustability that people need to know what's it's not a black box, what's behind the scenes so that they can trust the result they get from the AI algorithms.

RT: So interesting, isn't it? Because, yeah, I mean, there's, there's a lot of promise, but yeah, it's also like how it gets implemented. I suppose there's a lot of work that still needs to be done there. So that's, that's really interesting. So, I wanted to move on to one of the technologies that you know all about really well, which is digital twins. And there will be people listening that don't really know much about digital twins, and I just wondered if you could walk us through what they are and how they're being used, and what difference they can really make to support a more sustainable and resilient infrastructure environment. So be great to hear from you as an expert in this field, and you know, it's very complex when you're trying to understand it.

YT: Sure, there's a lot of debate, actually, as to define what a digital twin is. A digital twin is widely considered, essentially a living digital replica of a physical asset. So whether there's a bridge, a building or it could be an entire city, what makes it powerful that it's not just a 3d model, is it's continuously updated with real time data from sensors, from inspections or simulations, and this will allow owners and operators to monitor the performance, test what if scenarios and also ultimately make better decisions. From a sustainability standpoint, digital twins can help optimise energy use, schedule predictive maintenance and extend the life of assets. And from a resilience standpoint, they can be used to simulate natural disasters and emulate how infrastructure might respond to those disasters, which is crucial for future proofing our infrastructure.

RT: That's incredible, isn't it? I mean, to be able to run scenarios and see how it's going to affect buildings and infrastructure that's going to make such a difference to us. And is that technology quite different than to sort of what we've had previously? Because obviously, BIM has been quite a big thing over the years. So, does it give us more capability, or just a different type of capability than what we've seen previously?

YT: Definitely, because with BIM, right, BIM usually it was, it was developed during design phase and used a little bit of construction phase, and it is abandoned. It's not updated with digital twins, we are continuously updating during operations and maintenance phase of our infrastructure. So, it gives a lot more opportunities beyond that for testing scenarios and seeing how the infrastructure might react or respond.

RT: Do you think it's something that we're going to see sort of like, it's going to be like the best practice, this would just be a matter of course that we'll just have, you know, that's what every building or every infrastructure project will use will be these digital twins. It's good. It makes sense, doesn't it? I think, you know, to have it sort of joined up. And, you know, the ability to do simulations and really forward planning, is that what you see? Is that what you'd hope?

YT: Yeah, yeah. Definitely

RT: Excellent. It’d be interesting to see if it was, like, even on an individual basis, in the future that we can just see, like, what would happen to your own like, building project, you know, and your own house would be quite interesting. So, thinking ahead, where do you see AI heading next in construction and infrastructure? And do you think there are any major breakthroughs or shifts that we should be watching for?

YT: I think so. So, I see AI moving toward more integration and autonomy in our industry, as I mentioned earlier, right? We just use AI in isolated applications like defect detection or optimising construction schedules. I think the next frontier is to link in these together into more holistic and AI driven systems that can help us manage entire projects or even city infrastructure in a more coordinated way. And I think another shift will be the explainability making AI more transparent so that the decision makers can trust and act on its recommendations. And of course, generative AI is opening up new opportunities for design, automation, scenario testing and knowledge transfer in ways we are just beginning to explore.

RT: I mean, that feels so exciting. I'm sure, you know, sort of other stakeholders, like policy makers, industry, just generally, general managers, you know, people that might not understand it, you know the technicality of it, but the fact that that knowledge transfer can happen in a way that speaks to them, that is almost translated to what their needs are, I find that really fascinating. And, you know, the potential and what difference, and, yeah, the speed of change, you know, it's going to help with that, I would have thought, because I think often in every industry, you know, it's having the language of that. Group, and if AI can bridge the gap and sort of make it more like, you say, the explanation of, you know, of how these things are done, or what difference it's going to make, or anything like that, I think it can only be good. And I think that's really a very exciting place to be. I just wondered if there were any areas that might be flying under the radar. I mean, you know, you might have seen this. Are there any parts of civil engineering where you think AI could be transformative, but they're not really getting the attention just yet?

YT: Yes, yeah, operations and maintenance tend to get less attention than design and construction, but they actually represent most of the structures, life cycle, costs and environmental impact. So, AI can transform actually how we monitor ageing infrastructure, detect hidden problems and prioritise repairs. Another overlooked area is acuity. AI could help us design spaces and systems that better serve diverse communities and adapt to changing needs over the life cycle of our infrastructure.

RT: I think that's incredible. I mean, that's often when that sort of planners are looking, you know, sort of like, holistically, there's so many more needs we don't even know about at this present moment. You know, we might not have like, names for them, but, you know, just to see what we can do to support, you know, in the community, what sort of needs and like, how you can transform buildings to whether it's existing infrastructure, whether you can use those sort of tools, those AI tools, to help transform sort of infrastructure for those different needs in the community. I think would be quite interesting, you know, sort of see what it can start doing for, you know, different groups of people. Thank you. I think that's really interesting. You've given us a lot to think about. I want to shift gears slightly because you have a very exciting project that you've been working on, and you are the Co-editor of a new emerald journal, open access journal, called Artificial Intelligence for a Sustainable Built Environment. I wondered what inspired the launch of this journal right now. And what are you excited about?

YT: Yeah, because we all know AI research is advancing very rapidly, and sustainability is an urgent global priority, but there was not a dedicated venue that brought those two together in the context of the built environment. So, we hope that this journal will fill that gap, and it's meant to be a platform where engineers, computer scientists, urban planners and policy makers can share insights and shape the future of sustainable infrastructure.

RT: Fantastic. And so, the journal is now open for submissions. What kinds of contributions would you like to see come through. Is there anything you're hoping to attract?

YT: Right, so, we are looking for mix of rigorous, both technical and practical case studies that will include advances in AI methods, but also real-world applications that show impact. And personally, I would love to see cross disciplinary collaborations, where AI experts are teaming up with built environment researchers to address challenges in transportation, urban system, and energy systems.

RT: I mean, that'd be fantastic. And the fact that it's open access as well, that information can be read by anyone, so it will be really great for all the different stakeholders involved. You know, whether it's researchers or policy makers or industry or anyone that's interested in this field. I mean, obviously that will be a resource that they can use to, yeah, to find out what's going on right now. So, you know, that's one to watch. So, thank you. I look forward to seeing the first published articles. So, let's take a leap into the future. Now imagine it's 2050, what might a truly AI, enhanced, sustainable built environment look like in your imagination, with all the work that you do, and sometimes you can be very niche and it can be very focused, but when you sort of broaden that out, and you think about the future of what AI can do to our built environment, you know, what would be your hopes and dreams for that?

YT: I think there are big changes. I mean, 25 years for by 2050 I imagine cities and infrastructure that are truly adaptive, and systems that that can sense their environment and learn from it and adjust in real time. Buildings that can generate more energy than they actually use, and transportation networks perhaps self-optimise to reduce congestion and carbon emissions and infrastructure that can anticipate climate impacts before they happen. And more importantly, it's not just about technology, but it's also creating environments that are healthier, safer, and more equitable for the people who live in them.

RT: wonderful. And that's not very far away, is it? I mean, we think 25 years ago, it was sort of like, you know, 2000s which for me, just feels like just yesterday, but, you know, another 25 years, and that could be, you know, our future. That could be our environment. So, yeah, really exciting times ahead, and things will speed up by the sounds of it. I mean, the advances in AI, even just for, you know, a few years ago, you know, from what we see in the, you know, the general population, it's pretty incredible, really. And so finally, if there is one key idea or one takeaway, you'd like listeners to leave with today, what would it be?

YT: The key message I would like listeners to live with is that AI is not just a buzzword for construction, it's a tool that when used thoughtfully, can help us build a future that is more sustainable, resilient, and people centred. But that future depends on collaboration between engineers, AI, expert, policymakers, and communities, and together, we can shape AI to serve not just efficiency, but society and the planet.

RT: That brings us to the end of today's conversation on AI and sustainable infrastructure. A huge thank you to our guest, Dr Yelda Turkan, for sharing her insights. If you'd like to learn more about the topics we covered today, check out Artificial Intelligence for a Sustainable Built Environment, a new open access journal showcasing research at the intersection of AI, civil engineering and sustainability. You'll find more details and a full transcript of the episode on our website. And as always, thanks to our guest and to the team at this is distorted. This is Rebecca Torr signing off. See you next time you.