S1E6 - Navigating the Future of Transport with Real-Time Data With Nick Bromley

In this conversation, Aaron Phethean and Nick Bromley discuss the evolution and importance of transport data, particularly focusing on the integration of real-time data and mobile phone data into transport planning. They explore the challenges of data collection, the role of AI and big data in optimizing transport systems, and the future of transport data with an emphasis on privacy and security concerns.

Takeaways

• Good transport planning requires both short-term and long-term data analysis.
• Buses provide the most flexible capacity in public transport systems.
• Real-time data is crucial for understanding current transport demands.
• Historical data often fails to reflect current population movements.
• Mobile phone data can significantly enhance transport planning accuracy.
• Data collection methods must evolve to include modern technology.
• AI and big data can process vast amounts of transport data effectively.
• Privacy concerns must be addressed when using personal data for transport planning.
• Transparency in data usage is essential for public trust.
• The future of transport data relies on secure and anonymized data sharing.

Sound Bites

"What does good look like for transport data?"

"It's ludicrous to use 1920s data."

Chapters

00:00 Introduction to Innovation in Transport Data

03:28 The Importance of Buses in Urban Mobility

04:06 Understanding Transport Data Needs

06:54 The Role of Mobile Phone Data in Transport Planning

09:35 Challenges and Innovations in Data Collection

12:29 The Future of Data Privacy and Public Good

15:21 AI and Big Data in Transport Decision Making

18:15 Conclusions and Future Directions for Transport Data

What is Data Matas?

A show to explore all data matters.

From small to big, every company on the market, irrespective of their industry, is a data merchant. How they choose to keep, interrogate and understand their data is now mission critical. 30 years of spaghetti-tech, data tech debt, or rapid growth challenges are the reality in most companies.

Join Aaron Phethean, veteran intrapreneur-come-entrepreneur with hundreds of lived examples of wins and losses in the data space, as he embarques on a journey of discovering what matters most in data nowadays by speaking to other technologists and business leaders who tackle their own data challenges every day.

Learn from their mistakes and be inspired by their stories of how they've made their data make sense and work for them.

This podcast brought to you by Matatika - "Unlock the Insights in your Data"

Aaron Phethean: 00:15

Today, we are joined by Nick Bromley, a transport planning veteran and data innovator. We'll be exploring the links between company data and the analysis of company data, which tends to be on quite short 3 month cycles versus the extremely long term 5, 10 year planning cycles that is involved in transport data and people movement and very fine grained information and what's available. So let's dive in. Thanks for joining us. Hello, Nick.

Aaron Phethean: 00:44

Welcome to the show. I'm really excited for the discussion today. You know, we've worked together on several things and having a chance to sit back and discuss, that's that's quite a cool opportunity. So thank you. Thank you for coming on the show.

Aaron Phethean: 00:59

Yeah. Pleasure to be here. So I guess the the place to start is probably, you know, you are quite an innovator. You're into your data, particularly in the UK, and I'd even say you're a bit of an activist at the minute. You're trying to get things changed for the better for the for the country.

Aaron Phethean: 01:21

And, you know, you've had a quite stellar career across IT and transport data and using mobile phone data, one of the first to to make use of that. But, currently, it seems your passion is quite quite closely oriented with the bus network. So I suppose the question is why buses, and and why transport data?

Nick Bromley: 01:43

Well, I the reason why I like buses are I think it's the most flexible capacity we have in the system, whereas with trains, we're limited by, obviously, the tracks and the length of the station platforms and the signaling systems. Buses still offer a huge potential for growth if we can have the right vehicles running on the right routes.

Aaron Phethean: 02:07

And and I guess that's what it comes down to. It seems as though they're not really working because they don't have the right vehicles or they don't have the right routes. Is that is that a fair, summary of the situation? Well, yes.

Nick Bromley: 02:20

I mean, I as you know, Aaron, I used to work at Transport for London. And when I was working on big projects in the capital, I noticed fairly early on that a lot of the data we were using on buses was dating, in fact, back to the sort of 19 twenties, 19 thirties. So, yeah, that's if I got this right, I think 85% of the roots in London date back to sort of pre second World War. So, you know, I I sort of feel that the, the center of gravity of the population in London has probably moved significantly, but the question is, have we been able to properly predict where to put the buses?

Aaron Phethean: 03:00

Yeah. And that you you know, that I thought it was that throughout our time working together, that has been a continual point of fascination for me. It's like that length of planning time scale. I mean, it's it's kinda ludicrous. You know, we have this up to the minute information available spilling out of people's mobile phones and they're they're they're what they're doing through vehicle tracking.

Aaron Phethean: 03:22

It's it's there, but, yeah, the planning decisions are such long term huge infrastructure. You know, you know, it's a it's a big effort to put a new railway station in and that kind of time scales, and they're using 19 twenties data. It's just crazy.

Nick Bromley: 03:42

Well, yes. And in fact, even on the rail network, I, went to a conference back in, the early part of the summer, and it was clear then that even the planning of major infrastructure like HS 2, there are gaps in the data that you would anticipate would have been filled already, to deal with the demand and patterns of demand around their major cities.

Aaron Phethean: 04:06

Yeah. So we we have this relatively old data set to work with to help, you know, transport planners make these decisions. I guess and the obvious question is, what what does good look like? What would someone making a long term or short term, you know, transport decision ideally have?

Nick Bromley: 04:31

Well, I think you need to first of all split the time into 2. You've got real time, and then you've got building time series data, which could

Aaron Phethean: 04:41

be

Nick Bromley: 04:41

in days, weeks, months, years. So, basically, there are 2 categories to the data, but actually the same source. And so what good would look like is that every moving part of the machine, every person who's moving around in some way shares their data with either in real time or it's stored for future use in working out what the patterns of demand are, what the actual demand is, and what those things are gonna be. So that that would be good.

Aaron Phethean: 05:15

And I think the way you put it there is quite interesting. You know, the people moving, you know, from, you know, actuals, you know, origin to destination is is quite vital. I think one of the things that we discussed where the information that's available, someone going from one train station to another or one bus stop to another, it's not necessarily where they intend to go to. It's what they have to do, you know, because that's the nearest one to them. It might look quite different if you figured out the final destination as opposed to where they happen to get off the network.

Nick Bromley: 05:50

That's absolutely right. I mean, a bus station or a train station is is just a way point between your origin as you put it, which is probably your home address or your work or Mhmm. The shop you have to be in, and then the point you wanna get to. So, most of the, infrastructure where the modes all come together are just way points. So Yeah.

Nick Bromley: 06:13

That that's where the floor is in the system. We don't actually know where people are ending up.

Aaron Phethean: 06:19

So then, this probably nicely brings us to the the use of the mobile phone data that that you were, you know, spearheading and and one of the first to make available for these kinds of planning decisions. Can you tell us a little bit about, well, what the data can tell you? I I think there's a there's a wealth of information there, but to get your sense of what's what's contained within it for this scenario, and then what it was like to get access to that mobile phone data? You know, what was some of the hurdles you had to go through to get it to market?

Nick Bromley: 06:54

Yeah. Well, we first looked at this back in, 2007, when I first dropped well, fell into the problem, really. I fell over the problem, which was likely data, and we set up a small team to think about alternative data sets. We did that with what was Innovate UK, the technology strategy board at that time, and we had a working group, and I took away that some of the output thought further, and I thought, well, look, the one thing we're all carrying, that has some electronic value is going to be the mobile phone, the wallet, the wallet, the phone, and of course, all those are combined now. But at that time, the phone looked like a good thing to use.

Nick Bromley: 07:41

The problem was the very few devices actually had GPS data on it. So, there's a lack of accurate data. So, we looked at how we might use the mobile master network to triangulate much the same way a navigator on a yacht, that's my hobby, would look at 2 points, and work out what the bearing is between them. It's exactly the same thing. You look at a couple of masks, and you work out the bearing between them, and the point of intersection is fundamentally where that item is, or that person is at that point in time.

Nick Bromley: 08:16

Really, it came out of that, and then developing the algorithms to make judgments on whether people were walking on a motorbike, in a car, potentially on a bus. So, we'd look at the densities of points, and then we'd look at the speed of movement, and then we'd look at the type of routes that people were following. And out of that, we set up a business, and on the back of that, it was called CityLogic, which is now part of AMI. But it was always the first step, really. It was about starting to identify innovative, modern sets of data, and I think it was even then, we could see that the ideal was gonna be GPS data.

Aaron Phethean: 08:58

Mhmm.

Nick Bromley: 08:58

Mhmm. So that so that's where we are now really. It's how we get anonymized GPS data in, as I say, near real time.

Aaron Phethean: 09:07

Yeah. And that that I think, you know, we we look at a specific use case for a moment. If you were to investigate, you know, a big football event at Wembley, where has everyone come from? And that can tell you an awful lot about what you need to do to support that event, what the best routes would be, and it's, you know, it's an enormously challenging problem. There's a lot of data there, but also like you described, the way you arrive at it is not necessarily so, obvious in the data.

Aaron Phethean: 09:40

You know, if someone's walking versus on a motorbike and you're figuring out that through speed they're traveling, you know, that that's a kind of deep analysis at an individual's level to try and give you that information, you know, on the on the whole. And that that, you know, I suppose as a data problem, I find fascinating. You you have to work out how to collect high quality data and then you have to also work out how to make the the analysis, you know, to answer the questions. And if if we zoom forward then to today, where you've got this GPS data available, much, finer grain. You know, I think one of the issues is people are worried about how they would use it.

Aaron Phethean: 10:21

You know, I don't think any one of us would want our mobile phone tracked and made available to just anyone. So then I I think you've got some fairly unique ideas about how as an individual, you could control where it's going to, and perhaps even get some of the benefit of how they're using it. Do you want to share your thoughts on that for a minute?

Nick Bromley: 10:43

Well, yes. I mean, clear clearly, the data is already being used. It's just that it's not available for the public good, and it's being obviously collected by a combination of the device providers and the mobile network operators. And they have this pin sharp view of the world in terms of GPS data. But of course, it's only their particular population, you know, it's only their their sort of group of people that they obviously provide services to.

Nick Bromley: 11:13

So, they don't have a complete picture. So, even if they tried to analyze this, whilst it give you some good insights and better insights than probably we currently have, it's still an incomplete picture for what's really going on. You need to actually bring all of the, data together in one place. But of course, at the moment, they're making that data available to their developers. So, it's not like the data is not there, and it's not being used.

Nick Bromley: 11:36

It's just not being used necessarily for what we consider to be the public good and the sort of collective intelligence of what's going on.

Aaron Phethean: 11:45

Yeah.

Nick Bromley: 11:46

And, you know, and I share people's concerns that, you know, we wanna know what data is being collected, and who it's being shared with. You know, a nomination will be really important to this, and it's gotta be bulletproof. And the other thing is transparency. So where the data is being used and how it's being applied to solving these sorts of real world infrastructure problems, That needs to be completely transparent. So

Aaron Phethean: 12:11

Yeah.

Nick Bromley: 12:11

You know, that's that's key. But also, I I do see a sort of a progressive shift in this understanding and making this happen by things like the data use and access bill that's currently going through the House of Lords. So, it's non specific really to addressing this problem. It's sort of got itself, I think, slightly bogged down in the detail without seeing the bigger picture, which is essentially giving ourselves better public services in terms of our mobility.

Aaron Phethean: 12:40

Yeah. Yeah. Yeah. And I think, you know, the the the link for me to, you know, whether it's a company with a lot of data sources trying to make sense of it or transport decision makers trying to collect, you know, data from all sorts of, you know, people and, you know, companies moving around. Essentially, there's an awful lot of sources of data to form a complete picture, but there's also a huge upside if you can essentially allow many people to look at the data and analyze it and form decisions provided they know where it's come from.

Aaron Phethean: 13:20

So where what the quality of the data is or or, you know, also they could draw conclusions that are, you know, misleading perhaps. But, you know, you actually once you start to farm out that capability to people to make informed decisions on a complete dataset, well, that's when something truly innovative could could then arise. You know, why give just a couple of people that access and that benefit, and why have these incomplete datasets? You know, I suppose that that to me is the the bigger problem that's worth solving.

Nick Bromley: 13:54

Well, yes. I mean, but you see, we've we've had decades of of this very problem and how we've solved it because, you know, modeling and collection of transport data has been more of a science than guesswork probably since the mid to late 19 seventies. You know, a lot of a lot of modeling work in transport sector, so you can trace it back to sort of seventies and certainly the edges. So it's not like we haven't had 40 years at this. It's just that we've been using lots of samples of data, and we've been trying to sort of work out, you know, what we're looking at, a bit like sort of feeling an elephant in a dark room.

Nick Bromley: 14:33

You know, we sort of feel bits of it, and then draw conclusions as to what sort of beast it is. And that's become a whole industry in itself is what data are we going to use here and how reliable is it? And what's the modeling tool we're going to use here and how reliable is it? And these are becoming accepted norms, both within the public sector and transport planners, highway engineers, etcetera, but also the consultants who work for them. And, yeah, that's not a bad thing.

Nick Bromley: 15:03

Yeah. We had we had

Aaron Phethean: 15:04

to do

Nick Bromley: 15:04

something as I said because of the guesswork we were doing almost since pre war.

Aaron Phethean: 15:09

Yeah.

Nick Bromley: 15:10

But the world's again moved on. You know, there's, we've almost 50 years since a lot of this started to change. And now, the big change coming again.

Aaron Phethean: 15:20

And I think, you know, talking about what's coming. You know, I think it's now so at at one time it was impossible to have the full picture or even process the full picture. Yeah. Even if you had collected it, it would have been impossible. With AI and the investment in big data processing technologies, cloud infrastructure, it is possible to process it.

Aaron Phethean: 15:46

It is possible to have the complete picture. It might not be available yet, so we might not be collecting, you know, sensing, might actually have gathered the the full picture, but it is possible. Does that mean it's desirable? Like, do you see do you think AI and this kind of mass processing, do we need the full picture or is it kind of good enough to just stick with the and we're making an analysis of making an informed decision We're making a good enough decision. Do we need to have the full picture?

Nick Bromley: 16:19

Well, of course, uh-huh. That's sort of sort of catch 22. I mean, basically, you you need to re benchmark the data we already have out there by building a new model with a new set of data, and then you can see how wide to the market you are currently. I mean, it might turn out that certainly some parts of the network are pretty much optimized, but you might find there are some real disparities. But I'd also say you don't need to keep doing this.

Nick Bromley: 16:47

This isn't sort of a dashboard that you sort of sit there like some godlike being on a daily basis, reconfigure all the connections within the network and all the devices running over it, all the nodes. I think it's a sort of a one time, and then you let it settle down for another 5, 10 years. And frankly, the public purse can't afford to keep reengineering certainly the railways. But the bus networks, you know, there is quite a lot of flexibility and capacity that you could be retweaking fairly frequently.

Aaron Phethean: 17:19

Yeah. And that's that's the kind of typical time frame. It'll be a kind of 5, 10 year decision making process for transport infrastructure. Yes, it's a big decision. I think when you look at a company, they're no different in that you don't want to tweak and tune things every single day.

Aaron Phethean: 17:37

It might be reasonable you do that once a quarter or maybe do that once a year. You have high quality information to make a decision based on, you know, what you're what you're currently seeing and then and then head off. Yeah. There is a a time and a place for watching as you go and, you know, operational, you know, reporting and and and, you know, checking that buses are actually performing to timetables and and so on. But that's that's a different concern to this kind of long term planning.

Aaron Phethean: 18:08

I quite quite like that that insight into what what we should be doing with the data. So and take take us to the the kind of conclusion. What's what's the what next? What what do you think we should be doing and and perhaps wrap it up for us? What what do you think the future of transport data is?

Nick Bromley: 18:30

Well, I think any mobile device, be it satnav, Fitbit, you know, what phone, the data is automatically shared within some form of combined authority. But that authority is within the control of, you know, highly secure environment. So it's, you know, it's not available to everybody. So it won't be for the public just to sort of scoop up buckets of data for their own personal projects. This this data will be, you know, it'll be heavily guarded, highly secure, totally anonymized, and it'll be transparent then how you use that data.

Nick Bromley: 19:11

So when the public sector, you know, when a local authority decides to relook at the buses, then you'll look at those bus routes as they look at them in using the same data. So it's transparent that they're doing that work, and it's also available to you to also see how they're doing that work. So but I think the key the key thing will be to protect people's identity completely. That, you know, it's totally anonymized, and it won't be an opt out thing. Everybody's immediately opted in.

Nick Bromley: 19:43

And, you know, we've got precedent for this already, and if we look at the NHS and what they're trying to do there. So it's yeah. I I don't see another way of doing it, how you're ever gonna optimize and to make proper evidence based investment decisions unless you go down this route.

Aaron Phethean: 20:00

Yeah. It strikes me that whatever the data source, there's always this, concern about how it's going to be used. And, you know, as an individual, you have a concern about, you know, them tracking you, them, you know, the kind of people out there making, you know, decisions or being able to understand, you know, what what groups of people are doing. As as a company, you might have a more operational concern. Are people going to draw, wrong conclusions about, you know, our risk or our operations if I don't really understand what's going on.

Aaron Phethean: 20:36

So whatever the source, this this seems like a common problem. You know, kind of trying to solve that deep concern about how the data is eventually used. Yeah. Yeah. This is a this is a fascinating area.

Aaron Phethean: 20:51

I really, thank you for thank you for coming on, Nick, and, talking us through it, even at this high level. Really appreciate your time.

Nick Bromley: 20:58

Yeah. It's a pleasure. And, you know, I look forward to, doing some more of these podcasts with you.

Aaron Phethean: 21:04

Thank you. Excellent.