The industry currently has a major problem – we can’t measure recovery to normal service! Partly because we never agreed what ‘normal service’ is …
Picture this. It’s August 2022 and you’re trying to get home from work. Your train leaves King’s Cross at 6pm, but suddenly stops near Stevenage due to a major incident at Sandy. The train doesn’t move for two hours, whilst the incident is cleared, but then you’re back on your way. Your stop is Grantham, so you’ll only be late for dinner. Your fellow passengers from further north are in for a very long night!
But you’re just on one service, from just one operator. The East Coast Mainline is used by multiple passenger and freight operators. So how do we answer the question “when was normal service resumed?” and “did we do a good job recovering service?”. Below we explain why we think current measures aren’t fit for purpose to answer these questions, and how a new Location Performance Measure developed by JNCTION offers a radical new way to evaluate our performance.
Current measures don’t help us learn from journeys
- PPM and On-Time
- The rail industry has many measures which are helpful for long-term evaluation of performance, such as PPM and On-Time. But if we evaluate those for my journey in August 2022, we would see a drop off caused by delay and cancellations but have no way of knowing when services were flowing normally again.
- You might see an uptick as services start to be unaffected, but there would be a lag of several hours for these arriving at the incident location.
- Cancellations
- In major incidents, especially towards the end of day when operators have an eye to the next morning and having stock & crew in the right places, there can be high numbers of cancellations. So even if the Sandy area was clear, trains might still be cancelled at King’s Cross, or terminated short at Peterborough.
- On Time to 3, On-Time to 15
- In major disruption, services are often delayed by more than 15 minutes. Once you pass this threshold, the train will score 0 for all subsequent station stops including destination, even if it completes the rest of the service efficiently and with expected dwell section run times.
These measures are also currently done primarily on an operator level, mostly excluding freight. If LNER is showing an On Time improvement at 10pm, but Grand Central is not .. how would we treat this?
A clear picture of the service health of an incident location would help us fill in gaps in historic logs
What we really need to review our response to an incident, and fill in gaps in the historic logs, is a score for the incident location through time. By tracking when this score drops we can validate Incident Start Time. By tracking its recovery we can infer return to normal service.
This is what the Location Performance Measure does, as you can see from this example:
Read more about the R&D work which led to the Service Level Metric
A new metric can invite the industry to work together to bolster revenue, rather than finger-pointing to avoid delay fines
British rail is transitioning into a new period under the stewardship of GBR. As more and more operators are absorbed into the DfT Operator umbrella and combine forces to make larger teams following ‘best practice’ it makes sense to measure them on their contributions to a combined picture.
A fresh way to measure service levels encourages Network Rail, passenger operators and freight to consider themselves as contributing to a network. Once we remove the safety blanket of “we’re not paying for this one because it’s X incident type”, we can focus on what is really important: minimising disruption for our passengers and stopping them from churning.
With Location Performance Measure (LPM) data we will be able to give our maintenance teams the targeted training and support to help them be the best they can be. With Location Performance Measure data we can help our Control teams learn from incidents and act more decisively and confidently during disruption. We can stop throwing away revenue from having to pay the customers back for their journeys, or worse losing them from rail for good.
Read more about JNCTION Archive and JNCTION LiveData, which can help you better understand your historic performance and make better data-driven decisions.
To hear more, sign up to the webinar on 28th May 2025
Alex Clark, Head of Consulting
Further Information
If you’re interested in measuring your service recovery, benchmarking maintenance team performance, or providing best-in-class live service information to your passengers, contact Alex Clark at alex.clark@jnction.co.uk
To find out more about JNCTION’s products and services, visit www.jnction.uk
