Meet our digital difference makers: Andrew Bagnall

Sweco is Sweco thanks to its #DifferenceMakers. Our people bring the expertise, passion and diversity of thought that shape better outcomes for clients, communities and the environment. In this interview, Sweco’s Andrew Bagnall shares his journey into transport modelling, what keeps the discipline evolving, and why combining digital innovation with human insight is key to understanding – and shaping – the future of transport.

Tell us about yourself – and how you found your way into transport modelling

I started in transport planning as a student, and then as a graduate in 2000. I did fairly routine transport planning work for about a year or so, and then – quite fortunately – someone more senior I worked with said, “I think you’ve got a bit of an aptitude for analysis and modelling.”

They got me involved in modelling and I just took to it. I was hopefully pretty good at it, and I’ve basically been doing modelling and appraisal ever since. It really played into the way my mind works – problem‑solving, analysis, and I love a spreadsheet – so it felt like a natural fit.

What is it about transport modelling that keeps things interesting?

It’s probably a mix, but I’ve always liked innovation and developing new methods. Once I’ve figured out how to do something, I naturally start looking for how it could be done better – more efficiently, in a different way, or if it can be applied to a new challenge in a different setting or context.

I like finding a better way of doing something and then giving that to someone else so they can carry it forward. There’s always another challenge. And increasingly, digital tools are giving us many more possible solutions than we ever had before.

How do you explain what a transport modeller does?

We essentially use data to inform decision‑making. We try to replicate real‑world conditions in a model – basically on a computer – and then change things in that model to see what the impacts might be. That can then inform decisions.

That could be something as simple as changing the green time at a set of traffic lights and seeing what that means for drivers in terms of delay. Or it could be something really complicated, like building the Lower Thames Crossing and understanding what that does to society and the economy.

When should transport modelling be brought into a project?

Modelling should be considered at the earliest onset of a project. That doesn’t mean you have to start detailed modelling straight away, but you can plan it into the process.

It takes time to set up models, find existing ones, get data and understand requirements. There’s also guidance from bodies like the Department for Transport that major schemes typically require detailed appraisal and therefore some aspect of modelling. It really makes sense to bake that into the overall process.

How does transport modelling support broader disciplines?

One of the things we do is development planning – not just for a specific site, but across a whole area, usually a local authority.

For example, Sweco developed a model of Worcestershire, which we used to assess the South Worcestershire Local Plan, and we’re now using an updated version to assess the Bromsgrove Local Plan. We run the model in forecast mode, plug in what developments are going to look like – people, jobs, where they live – and then see where pressures emerge on the network.

It’s also about whether development is in the right place, because it’s not just about fixed land use – it’s about the relationship between land use and the transport.

Why does transport modelling matter in the bigger picture?

The bigger picture is that we’re trying to predict the future. And unless you try to make an educated prediction, you’re just making assumptions.

We’re increasingly moving towards vision‑led planning and then using modelling tools to assess what that vision actually means. Traditionally, models are based on empirical data – things that have already happened – so it’s a difficult path to navigate when you’re trying to model something that hasn’t happened yet. That’s where professional judgement and experience really come into play.

As expectations of transport investment broaden, modelling has a growing role in supporting value‑based decisions – combining evidence, judgement and an open view of uncertainty to inform where investment can do the most good, beyond simple financial metrics.

How has digitalisation changed what’s possible in transport modelling?

Compared to 20 or 25 years ago, we can now run a lot more scenarios and do it much faster. That allows us to explore a much wider envelope of outcomes.

There’s a real appetite from clients and funders – including the DfT and Transport Scotland – to see different futures and understand what they mean for schemes and development plans. Digital tools allow us to ‘supercharge’ optioneering at scale – not just testing one idea, but exploring many versions of the future side by side.

That’s where modelling stops being an exercise in simulation alone, and starts becoming an exploratory tool.

Where can AI add value in transport modelling?

AI has huge potential to support and develop transport modelling, particularly the ability to explore options quickly and at scale. It is already supporting us to develop tools and techniques to test scenarios, sensitivities and combinations that would simply have taken too long – or never been feasible – in the past.

At the same time, models in their own right won’t automatically predict the outcomes people hope for. For example, if an authority wants to switch 20% of people from cars to sustainable modes, there may not yet be an empirical basis for that. We can test interventions like road pricing and predict what might happen – but human behaviour doesn’t always respond in neat or expected ways, which unchallenged modelling and AI may suggest.

That’s where people remain essential. AI can open doors and expand the range of possibilities, but it still needs human memory, experience, judgement and curiosity to decide which of those futures are realistic or desirable. Used well, it therefore becomes a powerful partner rather than a ‘bolt-on’ tool.

What does that mean for the role of the modern transport modeller?

It’s about challenging yourself not to just keep doing the same old things. That’s true not just for modellers, but for anyone working in a technical or advisory role today.

We need to think about how new policies can be reflected in our tools and whether we need new techniques. There are approaches like activity‑based models that aim to model individual people and their choices, which in theory gives you more granularity.

But they’re very difficult to validate with the huge depth of data involved, and forecasting is even more challenging when you’re modelling things that haven’t happened before. The opportunity is huge – but only if digital capability is paired with professional judgement and a real understanding of how people actually behave.

How does Sweco support innovation and new ways of working?

Sweco is decentralised, and we’re happy to push efficiency – which really fits with my mindset. I’m a big believer in giving people agency: letting them go off, try things and do them better. That makes Sweco a good place to innovate and explore new ideas.

What excites you about working across borders and disciplines?

We meet regularly with modelling teams across Sweco – in the Nordics, the Netherlands, Germany and Belgium – and we’re increasingly working on international projects.

One project I’m working on at the moment is in the Philippines, where teams in the UK, India and Sweden are all contributing different expertise. That cross‑pollination of knowledge ideas really helps push digital thinking forward.

If you could go back in time, what advice would you give to the 20-year-old Andrew?

If you’re too tunnel‑visioned or siloed, you’ll struggle to make things better.

Be curious. Look outside your discipline. Learn how other specialists and other countries approach problems. One day you’ll have all sorts of magical digital tools at your fingertips, but always remember that they will be most powerful when they’re informed by your real‑world experience!

Finally, what excites you most about the future of transport modelling?

There’s a growing demand to explore multiple futures and understand what they mean. That’s where modelling, digital tools and human judgement really come together.

Digital gives us the ability to explore the art of the possible at a scale we’ve never had before. But I’m excited by the fact that it still takes people – their experience, curiosity and courage – to decide which possibilities are worth pursuing.