Navigating technological change in the evolving rail industry

In this episode, we're joined once again by Vaibhav Puri, RSSB's Director of Sector Strategy and Transformation.

Together, we'll explore some of the issues inherent in the adoption and implementation of new technology, our initiatives in this area, and some of the most promising new technological advancements.

Find out more at: https://www.rssb.co.uk/our-business-plan

Transcript

Host [00.17]: As technology continues to advance, opportunities abound. But it can be difficult to know what to adopt and how to go about it safely. And for the rail industry in particular, there are multiple other challenges that can make implementing new tech an overwhelming task. From questions around interoperability to cultural resistance and funding, introducing new technologies is no easy feat. We're committed to supporting industry in getting the most from the opportunities of new technology. 

With us today is Vaibhav Puri, our Director of Sector Strategy and Transformation. In this episode, we'll explore some of the issues inherent in the adoption and implementation of new technology. We'll also discuss our initiatives in this area, including the role of RSSB's Futures Lab, and highlight some of the most promising new technological advancements. 

Hello, Vaibhav, and thank you for joining us today.

Vaibhav Puri [01:14]: Sarah, thank you for having me. 

Host [01.16]: So to start, what do you think are the biggest challenges real companies face when trying to adopt new technologies? 

Vaibhav [01:23]: I think the biggest challenge the sector faces is around the short-term cost of change and the associated risk of successful implementation. Now, with technology, change is a big factor, but I think those two elements in particular are the ones that create the challenge. 

The key thing in my view is that there'll be adopters in sort of four main categories. You'll have innovators, you'll have early adopters, you'll have early and late majority, and then people who will lag behind in terms of what they're implementing. 

And if you think about new technology, particularly in this context -- because we are probably talking about something that the sector's unfamiliar with, or at least there isn't a widespread implementation already -- innovators will need support really at an early stage. Often, that comes with funding, there's a degree of research and other aspects involved. 

But fundamentally speaking, one of the biggest challenges that I have certainly noticed when working with people who are innovating -- or at least at that end of the early research levels, early readiness levels of a technology -- is that sometimes, there isn't a full understanding of the wider context of change and the associated challenges. Often, it's very technology-led because, unsurprisingly, those people are working towards and are championing a particular type of technology and are trying to champion a particular type of solution for the problems that the sector faces. 

And in that context, confidence really matters. They have to have the confidence, and we have to create an environment of confidence that allow things to be adopted and slowly implemented, and people learn while the implementation is occurring. 

Host [02.51]: Can you tell us how RSSB's Futures Lab is helping industry to overcome some of these challenges? 

Vaibhav [02.58]: I mean, the concept of Futures Lab -- and perhaps it's worth restating what the purpose was for RSSB to even set up this capability called Futures Lab. 

One of the elements here was not necessarily to look at something which is 100 years from now and really [has] no relevance of the railway of today, but really is to be prepared, to ensure that there was a mechanism in place to have the sector be prepared for a new technology that is coming along. So, a large proportion of RSSB will serve the customers here and now and will provide a lot of support. And then we need a capability that consistently looks forward and goes, OK, there's a new bit of tech coming around. How and what would be needed in order for that to be enabled? And if that is the case, how do we need to adapt our risk management approaches, our standards, to support that?

And a good example of that is the work in AI. So, know, AI has certainly become a huge hype, and it's being followed by lots of people. We kicked off a programme precisely to examine things like, how should you even think about an AI-powered solution? Is there a way of creating a common language that people can rely on so all the people can express their AI-powered solution in the same way, so we can really see what is working, what is not working?

And in that area, we have started to pull together a little toolkit, and we had a very successful workshop with the sector on the 18th of March this year to really showcase that toolkit a little bit and really learn. 

Another bit of that is thinking about the standards landscape that will affect AI, and that's a bit of work that we are also kicking off. 

Only other thing I should also highlight is some of our work around robotics and autonomous systems. So, I'm co-Chair of the Robotics and Autonomous Systems Standards, Ethics, and Regulatory Pathways Committee. So, it's a mouthful, I'm afraid. In that area, I think what we're examining at a transport level, not just at a rail level, and that's another thing to perhaps keep in mind, often for rail, the technology maybe sometimes coming from another sector or from outside rail. We are really involved in really helping the sector and the government and the Departments for Science and Tech to really understand what are the regulatory barriers in terms of implementation of robotics and the autonomous systems. 

Host [05.01]: Thank you, Vaibhav. Our 2024/25 Annual Business Plan outlined several initiatives around adopting new technology. Could you please provide a couple of examples, detail their progress, and give us an update on their current status? 

Vaibhav [05.17]: I think this is almost in some ways easier because I've started talking about it in the last question around AI. 

And the three things we identified in the AI programme in particular as an enabling organisation where, A, is there a common way of approaching AI solutions? And as I said before, we have sort of created a toolkit, an early toolkit, which aligns with the change management approaches that are applied in the railway sector, so around something called the Common Safety Methods and Risk Evaluation and Assessment. We've tried to align the AI approach to that, and that hopefully would work. And we unveiled that to the sector to get some early feedback. There's still some refinements to be done, and over the course of the next 6 months, we will work with individual entities to really trial the toolkit and really refine it further. So, that's one example of where we've made good progress and now more work to be done.

The second, in the same AI programme work, is one of the things that came out quite unsurprisingly when we talked to people was around human factors and the impact on people of the technology, like AI, or it would apply to anything else. Our human factors specialists really looked at good practice in other sectors and in other areas to really pull in, what is the design of an AI-powered solution with human factors involvement? What does that look like? What are some good principles that can be followed that we have learned from  automation, for example, that could be applied in an area of AI. And they've tried to embed all that good practice into the toolkit, but also we will produce [an] almost principles guidance document that will come out of that. So, that's still being refined, but in the next few months should come out. 

And last but not least, we have been working with BSI on a standards landscape for AI. And again, the approach here is to give the sector a mechanism through which they can navigate the standards landscape. I think that's the key. Our role here is not just doing this work, but then making sure that the outputs are easy to use for people. I think that's absolutely critical. 

Host [07.06]: And with the launch of our 2025/26 Annual Business Plan imminent, can you outline our future direction in this area? 

Vaibhav [07.15]: If I piggyback on the AI work as well, I think one of the things would be to again, like I said, work further and apply it on particular case studies and really refine what we have created already. 

But the other link there is the robotics angle here. So, if you look at AI, for example, in isolation, obviously we're doing some good work already and we will continue to build on it. But one key aspect of it is really thinking about robotics and with smart machines and really [starting] to think what does that mean? So, there's a case of sort of digital AI versus more sort of physically embedded AI. You know, what does that mean? And really start to understand that a little bit. That might be an area we explore further. 

Also really [starting] to think about validation and verification of AI models and solutions. That's an area that we are touching on at a very high level, but we might need to go deeper because as we start to examine use cases, there may be some specialisms, let's say AI's application in traffic management or AI's application in the management of crew, for example. There may be specialist areas that may require their own type of testing and acceptance, and maybe we can explore that in the future. 

And perhaps another thing that I should really plug here is the work we're doing on foresight and horizon scanning. Because if you think about innovation, it's really, you need to prepare the industry first in examining and really understanding what the new challenges are. And in a lot of those areas, technology, opportunity, and commitment kind of come together. Right? So, now we can help with the technology and the opportunity bit. The commitment bit comes from the sector itself. And so it's essential to have those mechanisms for foresight and horizon scanning work to be in place. So, the sector is consistently looking to the future to really understand what is coming down the chain. What are the sorts of problems I could possibly have? What emerging risks are there that could affect the sector? And the moment that problem statement becomes concrete, then the opportunity for technology increases because you're able to then say, to resolve that problem, I think there's also a technological development that might be helpful or an existing piece of technology that might be helpful. 

In this case, I think it's also essential to think about, when we talk about technology, there may be lots of existing bits of tech that can be applied in a completely new context. But the context has to unravel because we have to make the case for [the] problem and how we resolve it rather than the technology itself. 

Host [09.27]: Looking further ahead, what do you think are some of the most promising new technologies that could transform the industry? 

Vaibhav [09.35]: That's a good question because I think it's almost appealing to come up with something that looks, you know, like space-related that might excite people a lot more. 

But I think we have to stay grounded. I think we have to stay grounded in the sense that the rail industry has a lot of challenges today, and we have to make sure that we use these really interesting techniques and these really interesting capabilities in resolving those problems of today. And in the 5-, 10-, 20-year horizon, you can really explore the possibilities of applying new technologies to improving. Almost the use case identification becomes a bigger challenge than the technology itself. Because often the technology is there. The question really becomes, what problem are you trying to solve? What problem fits this technology best? And really making people aware of that and really pushing that aspect in so that people can be creative, the supply chain, the manufacturing sector can be creative and come up with solutions. While our job really becomes really shining the light on the kinds of problems that can be solved and fit between those problems and the technology that is essentially really coming across. 

Quantum is the classic one. You know, this is the year of quantum, I think. You could say quantum, what an interesting area, you how on earth could it apply to the railway? Well, there are bits. The quantum computing aspect is becoming more and more interesting. If you combine that with AI, and I talked about robotics, you know, only your imagination is the barrier. But also things like, I've certainly looked at labs and stuff around [the] sort of quantum based-signalling aspect where, you know, there's a possibility of new tech that can completely transform the way we identify where something is located. 

Another big area I think will be traction power for the railway in particular. Electrification is certainly there, but really, battery technology, hydrogen, and all these things are there. 

But we need to think very carefully about what are the areas where we can add most value. 

Host [11.21]: That leads me on to thinking about the testing and validation process. Is there work to be done here, do you think? And why is it important that these processes are as efficient as possible? 

Vaibhav [11.32]: It's a huge area in itself, testing and validation. And if we want to enable new technology, if we want to enable us to leverage the opportunities that come with new technology as it starts to bear fruit, we have to provide the ability to test and trial and really prove the concepts that are being, I guess, hyped up because of their useful application in other contexts. And that's an area that we have to be really, really prepared for. There's a lot of work to be done there in the sense of thinking about what kind of environments are needed, what kind of testing facilities are needed.  

You know, I hear the concept of sandboxes. Often, people will talk about we should have sandboxes. And again, you know, that's a word that's quite loaded. You can have regulatory sandboxes, which are much more about how to overcome regulatory challenges and are there any barriers there. You could have compliance sandboxes, which are all about what standards do I need to apply and are there standards that will stop me from innovating. And then you can have more sort of physical testing and trialling sandboxes where people are able to really test the idea that they have and really try and refine it and make it more usable. 

But the key thing in the testing and validation process for us has to be around if we create those environments -- and often they require a degree of investment, they're not cheap, they're not easy to get access to -- they have to reduce the effort that is required later on. And there has to be a degree of cross-acceptance as a result of that testing. What shouldn't be happening is that a technology has been proved in a particular context, has been implemented in a particular context, and you have a very similar application context, and people are insisting that it has to be retested. Because it completely undermines the whole concept of creating such environments because the whole idea is to reduce the effort. 

Host [13.10]: Thanks, Vaibhav. And finally, based on our conversation today, what advice would you give to rail organisations navigating the adoption of new tech? 

Vaibhav [13.10]: I think, very simply, think about the change. It's not about the technology. I mean, the technology is a key factor. We have to always step back and say, why do we talk about technology? Because what tends to happen is the technology has been applied. People can see conceptually how they can unlock certain things. Right? Otherwise it wouldn't be even under consideration. People can see it. There's an aspect of how do you actually deploy it. 

And in order to do that, you have to think about change. Change agnostic of the technology, but then seeing how the technology can be introduced in that chain, how that can be managed, and really thinking about the wider picture. That requires really thinking about the people at the end of the day. People insertion points and the pace of change. I think those are the sort of key things.

Perhaps, like I said, use organisations like ours to support and fill the gap.

Host [14.04]: That's a really interesting note to land on, Vaibhav. Thank you so much for joining us today. And thank you for listening. 

If you want to learn more about how RSSB is supporting industry in adopting new technology, please visit our website. 

We look forward to you joining us for the next episode. And in the meantime,  safe travels.