How Do You Solve A Problem Like Southeastern?


“Our train crew have been issued with Blackberry devices which allows us to provide them with more frequent updates, which they in turn can pass on to customers. Our website has been upgraded and is now able to cope with much higher traffic volumes [..] During the recent severe weather, it stayed available at all times. In response to requests from customers we have added a summary service information board to our website homepage which provides an overview of how services are operating.”

– Charles Horton, MD of Southeastern in a letter to Caroline Pidgeon AM

“The TOCs must recognise and accept that virtually complete dependence on the railway’s own electronic systems for keeping passengers informed in-journey during periods of disruption is just not acceptable. And the lack of any real alternative has been profoundly frustrating for very large numbers of passengers in the recent disruption.”

– David Quarmby, “The Resilience of England’s Transport Systems in December 2010”

With a month now passed since the major snow disruption of November/December 2010, some real evaluation of the disruption can now take place. Virtually all of London’s Transport Operators (TOCs) found themselves hit hard in some regard by the snow and ice at the time, but it was Southeastern who, by the middle of December, had emerged as the dubious poster child for the failures of various Train Operaing Companies (TOCs), particularly south of the river.

In addition, the nature of Southeastern’s failure highlighted clearly the issues with a TOC strategy that has become increasingly prevalent in recent times – a reliance on electronic means to communicate with passengers in all situations.

Trouble at Track Level

Running trains south of the Thames in severe weather conditions is inherently difficult. Put simply, much of South London’s Railway infrastructure suffers from that fatal flaw of history – being a technological pioneer.

In this particular instance, the technology in question is that relating to the transmission of power to the trains themselves.

In contrast to the overhead wires that you’ll see on the majority of Britain’s electrified railway network, much of South London uses a third rail system to deliver power to its units. In its simplest terms this means that power is delivered via a seperate conductor rail and picked up via a “shoe” on the train itself.

It’s a simple system and one that, broadly speaking, is relatively effective (although it does have its flaws – notably it makes it tricky to run trains close to or over 100mph). In snow and ice, however, third rail is a system that can be notoriously prone to failures.

If snow or ice forms on the conductor rail, then it causes problems in the transmission of power between rail and rolling stock. Indeed in many cases the newer the train, the greater the problem, as modern units are often far more sensitive to the power spiking issues this produces than their simpler predecessors.

Most modern third rail systems (such as that found on the DLR) minimise the ice issue by running “side” or “bottom” systems, where the train’s shoe avoids making contact with the conductor rail on the top (where snow and ice naturally settle). In South London though, the older “top” system is in place and thus the only real remedy is to prevent snow and ice forming in the first place.

Broadly speaking there are three ways this is normally achieved:

1) Heating the conductor rail
2) Running “Ghost Trains” through the night (empty trains that shift any snow and ice from the conductor before it becomes too much of an impediment)
3) Using Multi-Purpose Vehicles (MPVs) – rail units designed for heavy work – configured as anti-icing units to keep both running and conductor rails free of heavier snow and ice build ups than ghost trains can handle.

Unfortunately for both Southeastern and other operators, the three methods above proved only partially successful at the end of last year.

On Southeastern’s network, heated conductor rails are not currently fitted everywhere, and thus could only be of use where present. The Operator also attempted to run ghost trains through the night to minimise ice build up, but the severity of the weather meant these achieved limited success, and had to be curtailed in some places to avoid the risk of empty trains succumbing to the very power problems they were intended to prevent.

Finally, both the TOCs and Network Rail – the company responsible for maintaining Britain’s railway infrastructure – were caught out by mother nature herself. Arriving on the 30th of November, the heavy snow started far earlier in the year than had been forecast in any of the companies’ heavy weather plans. Network Rail’s MPV fleet was still largely configured to deal with keeping lines free of leaves, and wasn’t scheduled to be fully converted to an anti-icing fleet until the 10th December. Although this ultimately proved less of an issue for Southeastern than it could have (Network Rail claim that they met the vast majority of Southeastern’s anti-icing service requests) it was a problem nonetheless.

To make matters worse, many of the recommendations resulting from the snow issues at the beginning of 2010 (more heated conductor rails, fitting better de-icing gear to a portion of Southeastern’s rolling stock fleet) were not due for completion until either the end of 2010 or into early 2011 – the early arrival of the snow pre-empted them by several months.

A Failure To Communicate

All the issues above obviously contributed to Southeastern’s problems, and arguably many of them were outside of the Operator’s direct control. In some ways, however, it was not these issues themselves that would cause Southeastern the most trouble over the period of worst disruption – it would be the consequences of actions taken by the operator to try and mitigate them.

For on the 30th of November, knowing all of the above and with upwards of ten inches of snow forecast for parts of London, Southeastern decided to implement a lower-service contingency timetable.

Contingency timetables (or at least better, more consistently planned ones) were another suggested improvement that came out of this year’s report on the disruption caused by snow last winter. Indeed Southeastern were not the only operator to turn to a contingency timetable, in their case, however, there was one issue – and that was the decision’s timing.

Automation Everwhere

Another recommendation that had come out the TOC’s investigations into last winter’s snow disruptions had been that contingency timetables be implemented through a new computer system – the Integrated Train Planning System (ITPS). In theory this would make the various operators’ contingency timetables more consistent and more reliable. In another clever move (on paper at least) the ITPS was also hooked directly up to the railway’s Customer Information System (CIS).

The CIS powers the timetables behind an overwhelming percentage of the information services the TOCs provide – including, importantly, the various National Rail information systems and the information boards at stations. It’s a complex system and there are some serious constraints about what data can be changed, and when it must be changed by. In fact, if you want to change your timetables using ITPS/CIS you must do so by 5pm the day before if you want everything to run smoothly and automatically.

Here was where Southeastern’s problems arose – for by the time they took the final decision to move to their contingency timetable it was no longer the 30th of November – it was the early hours of the 1st December. Well after the previous day’s 5pm deadline had passed.

With the deadline passed, the only way the contingency timetable could be implemented was manually. This is an incredibly labour-intensive task, and it is to the credit of Southeastern’s staff at ground level that it was accomplished with any degree of success at all. Crucially, however, there is currently one major problem with a manual implementation – you can’t feed that information back into CIS.

Suddenly, starkly, just how reliant Southeastern were on providing customer information automatically was brutally exposed.

Empty information boards at Charing Cross, via 853, photo credit to @dollyrockaUK

To be fair to the Operator, it was not a problem exclusive to Southeastern – all the TOCs have the same reliance – but whilst other TOCs suffered later in the day as their timetables in the CIS became outdated, it was Southeastern who, from the moment the day’s operations began, felt the full force of the problem.

With no information in CIS, the boards were empty. The electronic displays at almost every location were either empty or out of date. Nearly all information online was also either incorrect or out of date and Southeastern found themselves having to ask National Rail Enquiries to turn off any information relating to their operations. The Operator may well have made the correct decision in moving to a contingency timetable but that mattered little to passengers standing on freezing platforms bereft of any information about what was running or – worse – getting information that was either out of date or incorrect.

Such a communication crisis could only have one solution – a human one. Here, however, Southeastern swiftly discovered that whilst shifting from highly-staffed operations to a more centralized, electronic approach is good for the bottom line, it is not particularly useful when you need bodies on the ground and out there communicating fast.

Southeastern’s staff suddenly found themselves having to act as virtually the only conduit of all information to passengers whilst also being tasked with keeping the lines running at all. To make matters worse, Southeastern’s internal communcations (again, like most TOCs) also relied heavily on the electronic information it now had to cover for.

Southeastern’s management may well have looked hopefully towards the end of the day, hoping that once the CIS updates resumed normally their problems would at least reduce to the (still considerable) issues of keeping the trains running. It was, however, not to be.

On Thursday, for reasons outside of Southeastern’s control (previously undiagnosed issues with the new ITPS which was, it is worth remembering, being used “in anger” for the first time), they suddenly found that all their station information had been overwritten by an unexpectedly late download. Attempts to fix this didn’t work and thus on Friday the service information was corrupted for various parts of Southeastern’s network (and others in the south), with normal and contingency services shown as running at the same time.

Thus for several days, Southeastern found themselves in the middle of a communications crisis that massively inconvenienced their passengers. Southeastern’s efforts to minimise the public relations damage on the wider stage also fell flat. Several attempted efforts at corporate-level damage control with broadcasters and newspapers accomplished the very opposite, when PR officers and Managers were exposed as being out of touch with the situation on the ground, and unable to provide information or answers that was actually of use.

A Game of Two Halves

Overall, on the infrastructure side, many of the Operator’s problems can reasonably be said to have been outside of their direct control. This is something that Southeastern themselves were keen to highlight during the disruption itself, and have continued to point out subsequently.

In focusing on the physical issues, however, Southeastern are, consciously or not, only addressing half of the problem. For a journey by rail is made up of two halves, of which it is the second – not the first – that takes place on the train.

The first part of any journey is actually being able to be in the right place, at the right time, to catch the train. It was Southeastern’s failure to put its passengers in a position to accomplish this – or even be able to decide whether they should attempt to travel at all – that was as big a failure during the snow disruptions as anything caused by physical issues.

That the failure to communicate effectively was a major problem this winter – notably for Southeastern, but for the other TOCs as well – is something that was not lost on David Quarmby, who was once again tasked with auditing the response of England’s Transport system to this winter’s difficulties.

As he rightly points out, the industry should not look on the new ITPS system, with its CIS links, as a suitable solution to all its problems during disruption. In fact, it should seem obvious that the system has a major limitation in its need for accurate, early data – the guaranteed presence of which is a ludicrous thing to rely on in any situation that can be defined as “disruptive.”

More importantly, Quarmby’s report also acknowledges that the idea of electronic systems as a communications panacea is something that too many amongst the TOCs seem to now subscribe to. It’s a dangerous belief to hold, for these systems are not infallible and when they do fail it is the passenger experience that suffers:

The fact that [disruption] is taken very seriously by the rail industry, and is the subject of major projects under the leadership of NTF, is no comfort to rail passengers stranded on platforms with no information about what is going on.

Southeastern’s failure here may have been the most obvious in recent times, but it is by no means an isolated one. Regular commuters on lines throughout London (and indeed the country at large) can attest to the problems this electronic over-reliance causes on a daily basis. Be it overhead line problems at Hackney Downs,or signalling failure at Lewisham, too often such issues leave passengers looking at information boards with nothing useful to say, and leave ground-level staff struggling to provide accurate information to everyone at once in additional to other duties – that is if the station in question is staffed at all.

As Quarmby rightly suggests, in some ways this has become a cultural problem at TOCs like Southeastern and it is one the industry still fails to acknowledge and address:

Whatever is done to improve the electronic systems for passenger information – and these are impressive and welcome – I believe there is a more fundamental point. There does not seem to be a cultural acceptance within the railway of the responsibility to constantly see a disruption situation from the viewpoint of the passengers, and to act accordingly to make the effort to keep passengers properly and constantly informed.

Within the railway there is too much reliance on the electronic systems – and an apparent feeling of helplessness when the systems are not functioning (for reasons discussed above) or when the scale of disruption is such that the pre-programmed information on the systems cannot convey what needs to be said to delayed and frustrated passengers. There just is not enough flexibility.

I am quite clear that the more complex the systems for providing information electronically and automatically to passengers, the more there needs to be separate, simple backup arrangements which will work independently. Many TOCs provide BlackBerry or equivalent devices to station and train staff, which are used all the year round for keeping operational staff and those in contact with passengers briefed about disruption and service changes, whatever the cause. These and similar arrangements have been shown to be valuable. But they do not wholly meet the need at smaller stations where staffing is not continuous, or where booking office staff are busy attending to passengers specific queries or ticketing requirements. Nor can they easily deal with major disruptions where large numbers of people need to be informed and advised what to do.

Overall, whilst Quarmby’s points are made in reference to the weather disruptions, they certainly hold true for the many parts of the daily business of Britain’s railways as well.

His suggestion that there has been a cultural shift amongst railway operators away from appreciating the need to communicate with passengers in the bad times – as well as the good – is one that many regular passengers would certainly agree with. Making the TOCs themselves see that this is the case may, however, be a much harder task – especially as ultimately it is the targets set down in Franchise documents that the TOCs realistically speaking must meet, not those of the passengers themselves.

Indeed sadly, cynically, some may feel that the video below, which “downfalls” the recent weather disruptions may ultimately sum up some TOC’s priorities far better than this article does itself.

David Quarmby’s Winter Resilience report can be read in full here. A letter from Charles Horton to Caroline Pidgeon, explaining some of the disruption from Southeastern’s perspective can be found here. A good layman’s description of the third rail setup, and its issues, in South London can be found on the BBC website here.

Written by John Bull
John Bull is the Editor of London Reconnections. A transport journalist and historian, his writing often focuses on the political or strategic challenges facing London's transport network and beyond.