SaviRPM is a powerful simulation and performance modelling application developed by Risk Solutions and available to the rail industry. It enables rail performance teams to generate better understanding and management of performance and capacity risk, using cutting edge methods including: big data analysis, agent based modelling, and interactive visualisations.
Savi provides a powerful strategic decision tool helping train operating companies and Network Rail identify where there are potential risks to performance, and remove system fragility from timetables and service operation, creating a more robust and resilient system to improve service performance for passengers.
Savi is unique in two ways:
- it introduces typical incidents and delays to modelled services, providing more realistic interactions between services
- through interactive visualisations it allows users to identify risk and model potential ways to reduce risk and improve performance, to find effective solutions.
The application has been developed through the jointly funded RSSB and Network Rail Sandbox programme. Our rail industry partners GWR, Greater Anglia and Network Rail, directed development to create the most useful modelling features to help solve performance issues, which were then tested throughly through real-world case studies.
Here is a short introductory presentation from Jonathan describing what Savi can do.
Prior to 2020 UK passenger journeys had doubled over the previous 20 years and the high growth trend was expected to continue. With critical parts of the network running close to capacity during peak times, small delays could easily propagate and be difficult to recover. The industry was under increasing pressure to build a more reliable railway – to run more trains on time today, while improving the rail performance of tomorrow. Savi was conceived as a powerful tool to address this challenge.
When COVID-19 struck, and we experienced lockdown restrictions, passenger numbers plummeted. Operators were able to deliver much higher service performance with fewer trains. Now they began to consider how they could build back services that would maintain this performance when passenger numbers returned to pre-COVID levels. Our work with operators and Network Rail changed tack as SaviRPM was used to explore both pre-COVID performance issues AND how to return from COVID lockdown maintaining this better performance.
Real world case studies
Case study 1: Greater Anglia troublesome trains: we explored the role of ‘disruptive’ trains in poor rail performance
Case study 2: Greater Anglia significant performance drivers: we modelled features of the COVID lockdown experience, to see which changes were responsible for the significant performance improvements seen.
Case study 3: GWR drivers of reactionary delay: how much does reactionary delay reduce if the trains persistently causing reactionary delays are removed or fixed somehow?
Case study 4: GWR performance risk with additional timetabled services: can we add additional services into a future timetable without increasing risks to performance? If there is a risk to performance, where might this be and how can we mitigate and prepare for the new timetable?
Case study 5: Network Rail capacity and performance risk in new timetables: how do different timetable options compare in terms of network capacity risk, and where / how should the designs be modified to reduce risk of lateness?
More information about these case studies is available from
Feedback from our rail partners
The case studies provided robust evidence that SaviRPM is a valuable addition to the rail industry performance improvement tool kit. Users have particularly valued:
- the revealing insights the application delivers into how the railway system, a complex interacting system, is working
- the ability to explore the feasibility, robustness and resilience of timetables
- the ability to quantify the potential risks of a new timetable design
- the ability to explore proposed interventions designed to mitigate delay events, and test their resilience over a range of abnormal and disrupted running conditions
- the ability to explore and challenge commonly held assumptions about the causes of, and solutions to, poor performance
- the potential to test solutions in the modelling environment before committing to expensive trials
- the potential to explore observed changes in performance, such as recent significant improvements during Covid and understand the causes of these.
“the modelling has helped us work through a very complex system to understand the individual levers we have to use to deliver change” Marc Ware, Performance Manager, Greater Anglia
“this adds another dimension to our insights, it is lifting the lid on a dataset that we had, but didn’t quite know what it meant”… “the exploratory process was very positive … it was not about telling us how to do things but prompting thoughts and further exploration” Keith Palmer, Head of Performance & Planning, Greater Anglia
“the modelling adds credence to decision making – it helps justify and explain decisions and made the internal process of changing the timetable smoother” Simon Greenwood, Performance Analysis Manager, GWR
“a good and valuable tool for exploring where timetable capacity and performance risk lies”… “it’s good to work with a team that develops the modelling with you in a truly collaborative way” Richard Raine, Programme Manager (Timetable Performance), Network Rail
These benefits are made accessible to users by the comparative speed and ease with which the application can be set-up, run and the results explored. The application is specifically designed to aid collaboration; our rail partners have recognised that its value is derived from the discussion of the modelling results, shaped by the interactive visualisations, and the learning this delivers.
We would be delighted to hear from you if you are interested in finding out how Savi could help you understand the drivers of good performance. Please get in contact.