How transport systems are being future-proofed for a resilient, digital, low-carbon era

Today’s guests:

• Damian Garnham, national infrastructure, transport, and regional government leader, Deloitte Australia
• Andrew Pau, transport sector leader, government and public services industry, Deloitte Canada
• Kevin Kelly, principal, Deloitte Consulting LLP
• Anant Dinamani, principal, Deloitte Consulting LLP

This episode of Government’s Future Frontiers explores how global transportation systems are adapting and transforming for the future, drawing heavily from Deloitte’s Global Transportation Trends 2025 report.

Host Tanya Ott is in conversation with Damian Garnham, national sector leader for infrastructure, transport, and regional government at Deloitte Australia, Andrew Pau, transport sector leader for government and public services at Deloitte Canada, and Kevin Kelly, principal at Deloitte Consulting LLP, on how intensifying extreme weather events, aging infrastructure, and other unexpected disruptions are exposing vulnerabilities across transport networks and supply chains: “Transport systems face hazards from several directions, not only from extreme weather events […]. Unexpected interruptions can come from a number of sources, such as breakdowns, accidents, plus a growing challenge comes in the form of cyberthreats […].”

Future-proofing these systems involves developing the ability to anticipate disruptions while building resilience against them. In this regard, artificial intelligence is an enabler for transportation agencies seeking to “do more with less” amid constrained budgets and increasing system complexity.

Anant Dinamani, principal at Deloitte Consulting LLP, explains, “AI is not new to transportation. Our clients have been using different types and techniques of artificial intelligence for at least a decade, if not more. And, you know, simple things like counting cars, detecting incidents on roads, predicting what the price of tolling should be. There’s [an] opportunity with AI [to solve] multidimensional challenges, where you’re not just measuring the number of cars on [the] road, but you’re also then trying to understand why there are so many cars on the road.”

Commercially, despite electric vehicles and autonomous vehicles being leveraged for improved cost efficiency, reliability, and scalability, challenges exist. “People are scared about hacking […] I had a driver of an Uber that once told me that I’m never concerned about a human being worried about getting home because everybody wants to get home. An autonomous vehicle doesn’t have that motivation, so if it gets hacked, there’s that concern.”

Regardless of the many improvements to our transportation systems, the central challenge is ensuring the smooth movement of people and goods within and across countries. Join us, as we unpack how governments and industry leaders are addressing this challenge, and many more, by rethinking funding models, strengthening resilience, and preparing infrastructure for a more digital, low-carbon future of transport.

Tanya Ott: Transport is vital to everyone: It ensures a smooth-running planet and enables us to access friends, family, workplaces—and allows goods and products to be accessible to us. But it’s also much more than that. It enables us to see the world, experience different cultures, and meet new people. Right now, the global transportation networks, that we as a population rely on, face significant challenges.

Those challenges come as planners and policymakers struggle to integrate new innovative technologies and transportation modes that are emerging all the time. And if you add to that growing cyber vulnerabilities, antiquated technology, and more frequent and intense weather events, our aging network of roads, bridges, rail, runways, and transit lines is being subjected to extreme pressure.

I’m Tanya Ott, and this episode of Government’s Future Frontiers plots a course through these challenges as we take a journey through [Deloitte’s] Global Transportation Trends 2025 report. So, buckle up as we explore the rapidly changing ways in which people and freight are moving around the globe, and how public and private sectors are steering toward a more connected, sustainable future for transport.

From shifting travel habits to greener freight routes, and the rise of AI-powered transport systems, Global Transportation Trends 2025 is a snapshot of an industry in motion. We’ll be pulling out some of the standout findings: How resilience and reliability are becoming just as crucial as speed and efficiency, and what that means for everything from daily commutes to global logistics.

To help me navigate the twists and turns ahead, I’m joined by a team of transport experts. Kevin Kelly is a principal with Deloitte Consulting LLP’s government and public services practice.

Anant Dinamani is a principal with Deloitte Consulting LLP and a leader within Deloitte’s future of mobility, analytics, and solutions team.

Damian Garnham leads Deloitte Australia’s infrastructure, transport, and regional government sector, and Andrew Pau is Deloitte Canada’s transport sector leader for the government and public services. Let’s hit the road!

Kevin Kelly: There’s nothing in our lives that is not impacted by transportation—whether it’s the movement of goods and services, the movement of people, the ability for somebody to get to the physician, the ability for somebody to get to the grocery store in order to buy food. It impacts every aspect of our lives.

Ott: One of the big talking points this year is how transport networks can weather the storms ahead, literally. With extreme weather impacts intensifying, Deloitte believes the challenge is to make our roads, rails, and ports tougher, smarter, and cleaner, all at once. That means rethinking how we move freight, too.

This could involve the wider use of low-carbon fuels [and] digital tracking that trims emissions and swerves congestion. So, just how much of a challenge is extreme weather? Andrew Pau is with Deloitte Canada; in recent years, he’s seen more than his fair share of disruption dealt out by extreme events.

Andrew Pau: I can probably speak more closely to the west coast of Canada. But similar things have happened in the east coast. About, well, five years ago, we had the atmospheric river that took out both rail lines going to the port of Vancouver—that basically took the port out for the better part of three weeks.

Ott: An atmospheric river is a long, narrow band of air in the atmosphere that can carry huge amounts of water vapor transported from tropical or subtropical regions toward higher latitudes. When that moist air hits the mountains or cooler land, it rises, cools, and releases that moisture in a deluge of heavy rain or snow.

Pau: The port was essentially out of business for three weeks in particular, and it took a while to clear the backlog after that. The large fires we had a couple of years ago also brought a lot of disruption in rail—going in and out of our ports—both in Vancouver and in Prince Rupert. An ice storm or a rainstorm in Atlantic Canada took out the ports on the east coast. So, certainly, from a weather-resiliency perspective, I think it’d be fair to say that we’ve experienced enough events now in our Canadian supply chain to know that the likelihood of something happening is there. So, now, the question is: Can you anticipate and mitigate or remediate [such events] earlier?

Ott: Andrew’s examples show [just] how rapidly disruption can bring the transport system to a halt. So, what are some steps that can be taken to help us make these systems that all of us rely upon more resilient for the era we’re living in?

Because right now, transport systems face hazards from several directions, not only from extreme weather events like the ones Andrew has described. Unexpected interruptions can come from a number of sources, such as breakdowns, accidents, plus a more recent challenge comes in the form of cyberthreats because more and more of the ways we used to get around rely on some level of control from integrated and automated technology.

Damian Garnham is one of the main authors of Transportation Trends 2025.

Damian Garnham: At its core, a lot of the challenge is still [around] providing really effective and efficient services to the traveling public, and also keeping goods moving simply around the country. The mission isn’t much more complex than that. The challenge for transport agencies today, certainly in this part of the world, is the challenge of doing that more effectively and efficiently in terms of cost.

Dealing with the rise in, and introduction of, new technologies like artificial intelligence, but then also looking to do so in a way that drives down emissions and how you move goods and freight across the country; but then also, you know, how you deploy new technologies like electric vehicles across such a large area.

There’s a renewed focus on pathways to 2035 in terms of emissions reductions and the changes that industries are needing to make to achieve that—our transport [sector] included.

Ott: Looking ahead toward that 2035 target means plenty of planning for those unexpected events, but Andrew believes how to prepare for those events has changed over the last few years.

Pau: I’d say, five or six years ago, future-proofing a port meant a lot around resiliency, so whether that’s for sure, and improvements and strengthening, dealing with rising ocean levels and increasing heat and temperature. Believe it or not, that affects the quality of the rails. It actually softens them up a little bit and wears them down pretty fast.

So, as we have moved into probably the last three to five years around the energy transition, future-proofing ports has a lot to do about providing energy and fueling options [and] understanding the future market of fuels based on, you know, what transportation modes are coming.

Ott: We’ll look at the future of fuels being used shortly, but the shift toward resilience reflects a changing landscape when it comes to travel—not least because of a world-changing event just five years ago.

The COVID-19 pandemic changed a lot of behaviors: Urban centers across the globe that saw millions and millions of people commuting to and from work every day fell silent, as much of the world stayed home. And now, emerging into a post–COVID-19 world, Damian believes that some of those habits have stuck.

Garnham: The work-from-home dynamic—the hybrid working style—has remained a part of our business and environment. I think the other thing is that aviation is also critical to the mix in Australia. And certainly, since COVID-19, I think, we’re now seeing a lot more investment in the aviation sector again, but then also thinking about how that plays into sustainable fuels and some of those aspects given that aviation is a critical part of the equation here in Australia to as opposed to say long-distance rail in Europe.

Ott: Damian makes a point here about rail-travel growth in Europe. It’s true that the cost and speed of aviation are still currently more attractive, but rail travel is gaining ground, especially for those medium-to-long distances.

Meanwhile, EVs on the road are becoming more common. While still way back in the rearview mirrors on internal combustion engines, EVs made up about 18% of all new cars sold globally in 2023—up from 14% in 2022.¹ This year, it’s predicted that over 20 million EVs will be sold globally, pushing the market toward 20% to 25% of new car sales.²

Kevin Kelly believes that a further push will come, albeit not from the consumer side.

Kelly: I think it’s going to be more in the delivery world, where you have the commercial organizations. Look, I can go online, and I can order whatever I need for my kitchen, and 15 minutes later, it’ll show up at my door. That delivery van—that delivery vehicle—that’s where I think we’re going to see a greater and greater optimization of EV or electric vehicles.

And the reason that’s beneficial is because most of those go back to a hub. And they can have centralized charging and the [other] things that they need. They have all the infrastructure that they’re going to be able to build in that single location to maintain and drive these. What we have to do collectively is identify how we are going to charge that population of vehicles at a commiserate level than what we had previously been charging them, as the benefits from a commercial lens tend to move in the direction of a more EV-based delivery fleet.

I think that’s where we as an organization, we as communities, need to lean more into driving the revenue as it relates to that because they will not only be some of our greatest consumers, but they will be some of our greatest producers or greatest opportunity to drive revenue.

Pau: Fleet operators are two groups, I guess. There is the long-haul community, which I think is still trying to evaluate what are the technologies that they’re needing to invest in in the future—whether it is battery electric trucks or hydrogen-powered trucks. They really need to understand the duty cycle of these vehicles for the intended use case, versus the kind of last-mile, short distance, routine distance, which, in the near term, lends itself to battery electric haulage more so than others. Certainly, there’s still a role for biodiesel. And then when we look at the application of automation. You know [for] driverless trucks, again, fixed routes, predetermined routes, are going to be more ready for adoption than long routes.

Ott: The real kicker would be if those behemoths on the road that Andrew Pau is talking about, those massive trucks, could be run in much cleaner ways. Andrew explains that the research is well underway, specifically on the Class 8 trucks—the heaviest category of commercial trucks used on the road—that haul freight over long distances.

Pau: If you think about the things that leave a container ship that might not go onto a railcar, but onto the back of a truck, that size and scale of the truck is what we’re talking about there. And so we’ve been involved in a test bed where we’re, you know, running a couple of hydrogen trucks.

We’re early days with hydrogen, for sure. A lot of it is still in the proving-out stage.

Garnham: I think Australia, comparatively, is still in the infancy of its journey in electric vehicles. And again, I think some of that is connected to all parts of the system needing to move at the same time. So, charging infrastructure—and you know, for example, availability of vehicle models—is also a factor here.

Ott: Ensuring there are cars that drivers want to drive is also important. And there’s currently a supply-and-demand challenge in Australia, according to Damian.

Garnham: You know, in Australia, the highest-selling vehicle models are utility vehicles—sort of open-back SUV-type things—and the available models of those types of vehicles in electric form are very limited. It’s only just really starting to become available.

Some of the government policy now is very much targeted at getting into the next phase of that adoption. But you know, that coupled with the so-called range anxiety that can exist here, the public want to see the proof in the pudding before taking the plunge and making the transition.

Ott: Australia is well-suited as not only a consumer of sustainable fuels but also as a producer.

Garnham: We have the raw ingredients—the raw materials—that are required. So, as part of that pathway to decarbonize, the government’s making a very significant investment in that area. The statistics are that to get fully industrialized is a journey toward 2028 to 2029, to really get that happening at scale.

Ott: Keeping the transport routes moving smoothly, freight flowing from port to track and onto trucks, plus ensuring commuters and travelers are able to make hassle-free onward journeys on either road, rail, in the air, or on the water, is also vital. And in the coming years, maintaining the infrastructure is going to be not only a crucial part of the jigsaw, but also it’s going to face tougher challenges.

Here’s Anant and Kevin. They work closely with each other in all things AI and transport planning.

Anant Dinamani: When you think of some of the more risk-prone areas to hurricanes or floods or wildfires, how do you plan for that? What are the scenarios? So, that’s again something where there is a lot of work starting to happen.

Kelly: All the different studies that have been done on the infrastructure across the globe. And we’ll focus on the United States. Across the country, funding has not been kept up in order to support the maintenance of that infrastructure. So, the biggest challenge that transportation agencies have today is: How do they do more with less? Or how do they also increase the pool of money that they have in order to support that aging infrastructure?

Because the reality of it is: Fixing something you have, it’s not a ribbon-cutting ceremony. It’s the things that you have to do every single day to make lives successful. So, that continuous investment and making sure that those pools of money that we have available to us are stretched as far as they can, that’s where AI is going to be critically important.

But you’re going to have to invest simultaneously in AI and simultaneously in maintenance, in order for all of this to be true. Because not only can AI support that maintenance, but it can also support those new opportunities to make that new bridge, that new road you’re developing, enable the dollars associated with that to go a little bit farther in a time when the cost of steel, the cost of asphalt, the cost of people—all those things—are on the increase.

Ott: But help is at hand. We now have more data sets available to planners, developers, and engineers than ever before—data that can help to predict the frequency of events that can cause disruption to infrastructure.

Kelly: As the proliferation of information becomes available, we collectively attempt to leverage that to do our jobs better every day.

People in transit are no different than you and I. As the information becomes available, they’re trying to leverage that, so that, instead of, in 1960, I’m sure when they were doing planning and it was one [extreme event] in a 100 [years], it’s maybe one in five years, or maybe it’s even more frequent than that. So, the evolution of how we do that today happens naturally.

The problem is in the maintenance of the asset that was built by the individual when they did this planning, when it was one in 100 [years], that’s where the problem comes in. What do we need to do then? And that’s where the leveraging […] and what Anant was talking about is the visionary. We’re bringing assets with that visionary to enable those digital twins and AI, and the pairing of all those things together, to start changing the dynamic thinking today, where you start getting into the what-if analysis. And you start to hyper plan—I’ll call it, maybe I’m coining a new word—how different things are going to affect […] and what are the things that we need to do today for worst-case scenario planning. And you have to do that always within reason.

I think this hyper planning, or this what-if planning, has always been there. It’s just now we’re throwing more and more and more variables into the equation, and that’s where AI and digital twin and this modernized technology speed [up] that process. Because if I were going to do that in traditional or more manual ways, doing this is: a. potentially not within the art of the possible; and b. [it] would take forever if it were. You know, I can do that in a rapid or quick succession. That’s the value of the technology.

Ott: Retrofitting may be unglamorous, but the cost, when compared to building new networks, is much more affordable. That’s where intelligent management of what already exists can help out. Here’s Damian.

Garnham: The cost of building a new rail network or a new element of something—regardless of how sustainable it is—is astronomical now.

So, every gain you can get through intelligently managing the network and optimizing what you’ve got pays for itself many times.

Ott: Like every sector, transport strives to be efficient, and as Damian said, it’s all about doing more for less because the funding model for transportation is at a major inflection point.

The traditional fuel-tax model is increasingly less effective in the era of EVs, higher efficiency, and changing travel patterns. And, so, transportation agencies should diversify and innovate their funding and finance models now to create sustainable, resilient revenue streams, while also aligning those funding mechanisms with the broader mobility goals of decarbonization, congestion management, and modal shift.

Kelly: The challenge in the United States with trying to get all that to flow smoothly is ownership. Each of those things that you talked about—the roads, the transit agency, the light rail and buses, tolling—those are discrete organizations. And even though we, as consumers, might use all three of those modes in a single day, those are bifurcated or split two ways, three ways, or whatever.

And the challenge is that the funding model that comes in the United States from the federal government is just as split or just as separate as the organizations that support them.

Ott: Encouraging collaboration is trickier though, with so many agencies involved.

Kelly: Although everybody understands the altruistic benefit of coming together, it’s a virtual impossibility to make it true. So, until the point in time [when] that becomes a commercially viable thing to do, building some of those multimodal capabilities is going to be a challenge.

But somebody’s going to stand up to do it, and somebody will lean into that. Because, if you think about it, maps at some level today enable me to figure out the fastest path to get from point A to point B, and it will include transit; it will include tolling; it will include whatever methodologies a method is available to me; it will do it. And that’s probably the closest thing that we have to multimodal.

And I don’t want to overstate the [role of] collaboration [in] this, because, as I said, people in transport are core to the mission and they understand the value. So, people in transit, people in tolling, they come together to talk about the best way to solve that problem. But when it comes time to execute, it’s impossible because the funding’s not there, and I think we’ve got to find a better way to drive the funding from the top down that enables this cross-collaboration.

Ott: One of the biggest issues that planners face is whether to make the consumer pay a share, too. After all, they are the ones using the resource, but they already feel stretched to the limit of their ability to pay. So the aim is to look for a fair, sustainable solution.

Kelly: I don’t care how you cloak it—whether you call it a toll, whether you call it a congestion price—whatever you want to call it, it’s all just a derivative of a tax. It’s taking money out of the consumer’s pocket and putting it in somebody else’s account so that they can do something with it.

We have got to figure out a way to increase the tax in order to increase the available funds that are available to those municipalities to continue doing maintenance.

Waiting and waiting is not the mechanism to solve it. So, whether that is increasing sales tax or putting road-usage charging or congestion pricing, or more road tolls, you pick your methodology. Increase the money coming in so that you can increase the amount of funds available to support maintenance.

People need to understand how much it costs to do these things, how much it costs to maintain these, and I think that’s the first challenge.

Ott: And that’s where the planners and policy makers have a role to play. Anant picks up where Kevin left off.

Dinamani: The other opportunity is driving efficiency in overall planning and operations.

So instead of charging more, you just get more efficient in utilizing what you [already] have. But there’s also the secondary part to this, which is efficiency in terms of how much we build.

At the macro level, we spend US$450 billion a year to build and maintain roads. Roughly 60% to 70% of that is net new construction—all highways—even if you built 1% fewer roads and you drove throughput and improved utilization. Because a lot of the highway capacity was built to address peak demand. What if you shifted the demand by 1%? What if you reduced the build by 1%?

It’s complex. But that’s another way [you] potentially could go after it, which is, see if you could pivot some of the money away from construction into digitization. And we are seeing it, right? Some of the leading state departments of transportation we are working with and the visionary leaders that have given us an opportunity to work with them, the way they’re making money available is that they’re literally taking money out of construction.

Now they made a bet, right? Their point is that we can deliver more value to the communities by digitizing, using AI, and using technology than by fixing the next five potholes. It’s a hard decision to make, but you do need leaders that can take that step and say: “You know what? We are not going to fix these things. We are not going to construct this road. We’re going to try a new approach and see if that’s going to bring incremental value.”

Kelly: One of the things that Anant just touched on is key, and that is behavior—[the] behavior of the consumer. Many times, as you think about things, whether it was congestion pricing in New York […] what it’s [all] about is changing the way the assets are consumed, whether it’s a road, a bridge, or whatever.

The reason or the value proposition of these taxes isn’t necessarily many times to increase funds, although it does do that. It’s to change the behavioral characteristic of the individuals that are currently consuming that; move them potentially from more road usage to more transit usage.

And if we can do that, there’s an incremental value that creates more money for the maintenance; reduces our collective carbon footprint by having fewer vehicles in the area; makes them potentially safer by having fewer vehicles so that I have more pedestrians; increases the consumption of public transits so that those organizations today that are struggling financially get more people, higher ridership, [and] greater and greater benefit. And then I can start to leverage the assets that I have in and around a community, in and around a city for the purpose that they were built for.

Ott: In the short term, at least, consumers will also need to pay their way. But we haven’t mentioned the government yet. Damian says that, in Australia, the government is a critical stakeholder.

Garnham: It [the government] has the responsibility. Now, it owns the assets in many cases. There are some elements of the Australian network that are privatized, but not many.

I think a similar analogy is Canada or the United States, where we have the state and local and federal roles in the system. So it’s really important, I think, for each of the players there to work together, which is sort of a direct investment in a consistent and cohesive way.

But then, yes, [we should] be actually looking to maximize how far the investment dollar can go and the value for money they get from making some of these investments, but also trying to experiment in the right way with the private sector. So, to be able to try, you know, some of the emerging technologies in a safe way that allows them to scale and be adopted, if they’re proven.

Ott: When it comes to vehicles, the manufacturer has a major role to play, when bringing the new technology to the market. When that technology is ready to be rolled out, the government’s job is one of helping to scale the changes and put in place a legal framework to ensure safety.

And this is something Andrew Pau has been keen to tell me about.

Pau: These are complex value chains that are being redesigned a little bit. There are some regulatory changes that need to be made. There are operational safety components to be considered. There’s talent that needs to be considered; if you’re looking at new physical technologies, you’re looking sometimes at different supply chains.

These are nontrivial kinds of transitions. So, the manufacturers are investing R&D [research and development] dollars. Now, I think, our experience with this right now is the focus is on the regulatory landscape needed to support broader operations and adoption. What are the infrastructure investments that need to occur? On road, what are the things that communities need to be aware of? [For example] fire rescue, in order to have these machines safely operate in communities.

Ott: And while we’re talking about hauling goods around continents for distribution, it’s also a good time to talk about those vital freight terminals themselves—a sector largely overlooked until recently.

Pau: The sector has been ignored by the broad[er] public. COVID-19 brought it into the consciousness of the average citizen, if you will. And then, I think it was the Ever Given that had that incident in the Suez Canal; and then, now, with the water challenges in the Panama Canal, people now are having a much better appreciation of the importance of our supply chains and the people who live and work there in keeping our economies moving.

Ott: Of course, when we’re talking about new technology being applied to systems such as transportation, the biggest single change is, perhaps, the involvement and evolution of AI. Artificial intelligence is being used to make transport networks smarter, safer, and more efficient. AI can help by analyzing traffic patterns, predicting maintenance needs, and managing logistics, all in real time.

The sensors and cameras around the transport network, along with the data from connected vehicles, all help to improve traffic flow around cities. Reduce congestion at hotspots and, in turn, reduce emissions. And it works. Buses can speed through once-congested zones as they trigger traffic lights in their favor.

Travel patterns can also be examined, and better schedules and pricing modules implemented, all to encourage commuters onto public transit. Applied to freight, AI can improve fleet management, which can save both fuel and time. It can even forecast where there may be disruptions in supply chains and reroute where appropriate.

And while the usual challenges around trust remain, there is a potential for AI to act as a digital nervous system in the future of keeping networks slowing, thinking, learning, and adapting.

Dinamani: AI is not new to transportation. Our clients have been using different types and techniques of artificial intelligence for at least a decade, if not more.

And you know, simple things like counting cars, detecting incidents on roads, predicting what the price of tolling should be—so, there are many examples, right? The thing is [that] a lot of that AI was used to solve what we call single-dimensional challenges. And I’ll take tolling as an example:

It’s great. You’re essentially measuring the number of cars that are going on your road, and you’re pricing it according to that—saying if there are more cars, you’re going to charge more. Right? That’s a great example of what we call a single-dimensional use case. The opportunity with AI is solving multidimensional challenges, where you’re not just measuring the number of cars on the road, but you’re also then trying to understand why there are so many cars on the road.

How is that in relation to public transportation density, work hours, [or] events that might be happening in the city or in the region, [and] general incidents that might be taking place? The point is it’s endless. Same thing with safety, right? AI today is being used a little bit to understand where collisions are taking place, but then again, you could correlate to [where] drivers [are] braking hard. You could be correlating that to why people are walking or biking to begin with.

So my point is like, the opportunity with AI, the next generation of AI in transportation, is going to be about solving these multidimensional challenges, which brings me back to this network point—which is the number of government agencies and private sector [players] that figured out a way to share data in a collaborative, responsible way, while not compromising privacy, security, and trust in this whole process.

That’s what this is going to come down to, because the more and more data you bring, the more intelligent the system is. And the question is: Are we ready to embrace all of that intelligence from a human lens? Is human ingenuity ready to meet AI ingenuity? And that’s what this has come down to, which is: Will we operate differently around all this technology, and are we prepared for it? There’s no idea, but we can tell you what we are trying, right? Which is proving out the value of this collaboration one use case at a time. You’ve got to deliver value down to the community, because that’s how people have chosen to live.

Ott: Damian Garnham believes that AI can do what so far has not been possible: Study the whole of the transport network as it functions together—not simply as individual parts of a bigger puzzle—and it can identify how situations can have a knock-on effect from one form of transport to another.

Garnham: What’s happening day to day in a major city of several million people? It’s pretty dynamic. One incident that occurs can create major bottlenecks or a butterfly effect across the network, and that can cause a lot of disruption. So, where machine learning [and] AI technologies are really, really useful is to be able to take inputs from those situations and to simulate and predict some different outcome models to get the most effective solution or recommendation for people managing or running that network. So, there’s still very much a human in the loop in some of those scenarios, but they’ve already proven that out very significantly.

Similarly, operational planning or something like an airport, for example. If you know where the planes are, if you know the weather coming at you, a number of other variables, then the use of those technologies can really actually say, “Hey, if I adjust the speed of that aircraft coming in, I can avoid a disruption on the ground or I can either remove or minimize passenger delay.”

Ott: And what about helping deal with major hazards? Damian cites examples of wildfires when thinking about ways the IoT [Internet of Things] can play a role.

Garnham: You can actually deploy remote sensors in a bushfire area. I’m aware of some technologies here that will allow you to monitor a high-risk area or fire, for example, and effectively set up a network of sensors across that.

So you are feeding your digital twin with almost a pop-up of information that’s in an area that you might be concerned about. All these different inputs coming together, it’s not just about a sensor on a road, or information that the weather authorities hold, or some of the actual information that’s inside different modes of transport or different data sources that, before now, might have been very separate, but can come together and then have some serious processing power applied to them to start to predict and then, more importantly, be presented to a human with a set of decisions to make.

Ott: The scenario created by the use of such advanced technology and innovation is one of a smoothly functioning transport system. Something to look forward to? Well, maybe.

Dinamani: There’s so much excitement around the technology that you could get lost in the hype cycle.

Spend a lot of money, and take years before you see value. Therefore, applying AI to the most pressing challenges that these transportation agencies face, and proving out the value at that intersection and scaling from there, is generally the approach we are taking. Maintaining roads is a very expensive exercise.

Just in the United States, we spend about US$100 billion to US$150 billion, a year, maintaining roads. It’s a lot of money. The cost for maintenance has gone up because the cost of materials has gone up, labor has gone up, and traffic volumes have gone up. Weather patterns have reduced the life of the roads, and all these agencies are already strained for resources.

So, can AI really help extend the life of the roads? Can AI really help increase the reach of the maintenance dollars? And it’s not an easy problem, but that’s one very critical tactical example of kind of where we are seeing and applying AI. Another example is safety, right? In the United States, unfortunately, we have been losing more lives to roadway accidents and collisions.

It has been a huge priority for the United States to try and make our roads more safer. There are many states that have laid out Vision Zero objectives—which is essentially to get to zero accidents on their roads—and unfortunately, they’re very far from it. And every time we lose a life, it really impacts our confidence, our ability to get to that mission. Again, can AI help all the way from assessing the risks to improving the process through which we identify countermeasures, design ideas, bring visibility and prioritization to areas, intersections, neighborhoods, and communities in a faster way? Another example is around planning.

Communities are evolving very fast. The composition of communities is evolving fast, and the United States, historically, we have built roads to grow, but the challenge is that, in a lot of these areas that we want to grow, there’s no space. So, we do need to plan differently. Therefore, can AI help integrate land-use planning, housing coordination, transportation, and demand planning better?

Today, our transportation planning is sometimes five to 10 years behind real estate. It’s a huge gap. Can AI reduce that gap? Reducing that gap has value in itself. So, we are taking a very practical approach to these problems. We are trying to measure returns on investment at every step of the way [because] we’re trying to build confidence in the industry.

There is an opportunity here, but you have to take measured steps. But we are in an early, early stage of proving all of this out.

Ott: We’ve reached this stage in the podcast about the future of transport and trends, with barely a mention of autonomous vehicles. AVs are moving from experimentation toward integration and have the potential to reshape both passenger and freight transport.

AVs can reduce accidents, ease congestion, and increase accessibility, especially when combined with electrification and smart infrastructure. However, widespread adoption still faces multiple hurdles, such as regulation, public trust, safety verification, and the need for digital infrastructure that supports connected systems.

So, while AVs are still seen as a key part of [the] transport ecosystem of the future, their rollout will be gradual and uneven, which presents challenges in itself. But before that …

Dinamani: The technology has advanced very significantly in the last few years. It’s already proving out that it is safer than human beings driving.

In some cases, it’s also starting to make economic sense, especially in the ride-hailing part. People who used to previously rely on ride-hailing could potentially get a Waymo for cheaper. In the short term, in the delivery category, in the ride-hailing or ride-sharing category, there is a potential explosion of AVs here in the coming years. But would it still make sense from a consumer lens?

So, there are many questions yet to be answered in this space. We think there is a massive opportunity, especially from a safety lens. Is the opportunity beyond safety? Will it transform how we commute, and how we share, and how we live? And like beyond trust, the cybersecurity side of it, the technology side of it, right?

Like, people are scared about hacking, and people are concerned a little bit about … I had a driver of an Uber that once told me that I’m never concerned about a human being worried about getting home because everybody wants to get home. An autonomous vehicle doesn’t have that motivation, so if it gets hacked, there’s that concern.

Now, of course, security is paramount. Operators are obviously putting a lot of emphasis on it, but there’s also that concern of like: What if there was a loophole? What if connectivity were lost? What if communication were lost? So there are the trust factors. Beyond just the AV technology, it’s also in the security aspects of it.

Ott: Meanwhile, in Australia, Damian is acutely aware of the potential of AVs.

Garnham: One of the areas Australia—this is a private sector example—leads the way in this is the application in the mining sector. You know, there are autonomous vehicles in the heavy-vehicle space, massive trucks move all around from the mine, but those aren’t on transport networks. The technology is being applied there. The comment I would make is that I don’t think it’s an all-or-nothing scenario there. There are degrees of autonomy from none to fully autonomous, but the example that, certainly, the ITS Australia network, part of the Intelligent Transport Systems, promotes is actually where vehicles are able to communicate and take guidance or direction from roadside infrastructure, and there’s very much a safety case and argument there.

So, can a vehicle receive a signal from a piece of roadside infrastructure that there’s a hazard ahead or a pedestrian is in the thin middle of the road, and those kinds of things? There are trials happening here, which are testing that level of autonomy or connectivity between the vehicles. I think that’s probably where the real benefit lies. Certainly, there are examples in Europe, where you might enter a motorway and you might have certain constraints applied to you by the roadside infrastructure around you, but not be completely autonomous.

That safety angle is probably the version I think that most authorities here are looking at and [trying to answer] where they need to invest in roadside infrastructure to make that possible.

Ott: That’s a lot of discussion about road and rail transport, but what about heavy freight movements?

Pau: On the digital side, there is the ability to predict and manage and schedule, and slot vessels into berths. Certainly, our UK firm’s been involved in doing some of those things with some of the ports in Europe. Also, if you pointed to shore power, and certainly, there is a lot of exploration on that.

Ott: Shore power is when a container vessel or oil tanker can hook up to an electricity source in port so that they don’t have to keep their engines running while in dock to power the ship.

Pau: There are a couple of key barriers to figure out and overcome, like one, […] physical space for the electric infrastructure [being available], and then the second piece is the strength and resiliency of the grid itself. And to power a vessel that’s at the harbor with ground electricity, it requires significant amounts of generation to be transmitted over to a port. And of course, those transmission lines, those substation upgrades that need to happen, all take up room as well.

Ott: And how close are we to having self-driving trucks?

Kelly: Will there be a world without truck drivers? There probably is, and probably within certain sectors and certain geographies, certain type of haulage. But to say it across the board will be tough.

Pau: I think it comes back down to application again. I have a hard time seeing automation of trucking through the Rockies, let’s say, in North America. Just, given the weather conditions, the whiteness of the roads, the grades of the roads—that’s going to take a bit more work.

Ott: Damian is clearly focused on the fact that the future of transport is still based on one simple mission—a mission that is simple to imagine, but a bit trickier to implement.

Garnham: I would come back to the core mission that a lot of our clients have, of moving people and goods around safely and as frictionlessly as possible, if that’s a word. And there are still some very traditional challenges that lie at the heart of that, as I say, like cost escalation and inflation across the world that we’ve seen, rising citizen expectations in terms of use of technologies, and then also just doing more for less—that good old government challenge. So those are still with us and will remain part of the sector. I think, you know, always. But coupled with some of these emerging pieces in the trends that are hitting harder and harder as we go along.

Ott: Thanks for listening to Government’s Future Frontiers with me, Tanya Ott, brought to you by Deloitte Insights. I want to thank all of our guests today: Kevin Kelly, Anant Dinamani, Damian Garnham, and Andrew Pau.

If you’ve already subscribed, you’ll get new episodes delivered automatically. If you’re not subscribed, maybe hit that button right there, right now, so you don’t miss out.

This podcast is produced by Deloitte. The views and opinions expressed by podcast speakers and guests are solely their own and do not reflect the opinions of Deloitte. This podcast provides general information only and is not intended to constitute advice or services of any kind. For additional information about Deloitte, go to Deloitte.com/about.