Mark Phillips
Chief Executive, RSSB
Running a railway safely is more than just ensuring that every component—mechanical, digital, and human—works effectively.
The essence of system safety is that when these elements are brought together, they work in harmony as a whole.
Britain’s railways illustrate this well, with a wide and diverse range of interconnected parts.
The bogies and wheelsets of every train, locomotive, and wagon using the network must align precisely with the standard gauge of 4ft 8½ inches.
The speed, stopping positions, and headways of these trains are controlled by automatic signalling, to maximise capacity across the network and to ensure safety.
And at the heart of this ecosystem are the staff: drivers to operate the trains, signallers to oversee the flow, and engineers to maintain the equipment and infrastructure.
Effective functioning whole
These elements must all come together to form an effective functioning whole. To ensure these interdependencies hold, we need a discipline of system safety: standards.
Standards provide the connective tissue between components designed, built, and operated by dozens of different organisations.
RSSB is the body responsible for maintaining these standards, without which the system cannot function.
With them, operators, infrastructure managers, and suppliers can work to common expectations—and when something goes wrong, there is a shared framework for understanding why.
Most of Britain’s railways still depend heavily on people. Unlike fully-automated metro systems, conventional rail involves significant human participation at every level—from driving and dispatch to maintenance and control.
Ensuring that the right people are recruited, trained, and regularly assessed is therefore not a workforce issue alone: it is a system safety issue.
RSSB contributes here too, analysing how people carry out safety-critical tasks and making recommendations on training design, process change, and engineering adjustments where human factors create risk.
Data for safety
The evidence base for this work sits in Safe Insights, RSSB’s repository of safety incidents and events across the whole network.
From this data, RSSB has built a safety risk model and a precursor indicator model that members—passenger and freight operators and infrastructure managers alike—can use to manage the risks specific to their operations.
More recently, RSSB has begun developing predictive capability: forecasting safety events based on known risk mitigations and projected industry growth.
Artificial intelligence (AI) will accelerate this significantly. AI is already transforming how complex systems are modelled in other sectors, and rail is no exception. For RSSB, it offers the ability to model system-wide interactions in real time, identify where a change in one part of the network creates unintended consequences elsewhere, and better predict human behavioural risks before they materialise.
Tasks that were too costly to automate historically are also becoming viable—shifting human effort towards where judgement matters most.
The tools we use may become more sophisticated, embracing advances in technology, but the fundamentals remain the same. The parts must function effectively but so also must the whole; and standards will remain a key tool in ensuring this happens for the safe and efficient running of the railway.