We’re sorry to announce …, part 2

(… continued from We’re sorry to announce …, part 1, posted on 21 Jun 2016).

As appears to be generally the case, the inspiration for the high service reliability targets for HS2 that were voiced by HS2 Ltd Technical Director, Professor Andrew McNaughton, comes from the Shinkansen network in Japan. Indeed, the SYSTRA report acknowledges that “the on-time performance in Japan is exceptional” (see footnote 1). The report attributes this exceptional performance to be due partly to “how the Japanese train operators count the train delays”, but mostly to the Shinkansen being an “isolated line, without a mix of services running on high-speed and conventional infrastructure” (see footnote 2).

The report tells us that, in contrast “most of the European High Speed network is ‘open’, with trains running on the conventional network before and/or after their run on the high-speed lines” and that, as a consequence, “the performance of the European High Speed Lines is such that a significant percentage of trains incur delays of 10 to 15 minutes”. The exception that proves this rule is the Spanish High Speed network, which, the report informs, is “closed” – this being “due to a track gauge difference” which means that conventional trains on the rest of the network … cannot access”. According to SYSTRA, the exclusion of conventional trains from Spain’s hsr, along with “a relatively low level of traffic”, “chiefly explains its “very good on-time performance” (see footnote 3).

That HS2 will indeed be very “open” is evident from the indicative service specification for Phase 1 that I included in Degrading practices, part 7 (posted 12 Apr 2015). This diagram shows seven out of the ten peak train paths that are allocated as being utilised for classic-compatible services, representing 70 per cent of the allocation, or perhaps more fairly, 39 per cent of the 18 possible train paths.

An appreciation that there is a distinct risk that the service reliability of the legacy railway in the UK may exert a malign influence upon the performance of HS2 services, requires a realisation of what level of reliability is currently being achieved. Network Rail e-published statistics show Virgin Trains West Coast as achieving 91.4 per cent trains arriving on time over the period 1st May to 28th May 2016, and 86.3 per cent over the year up to 31st March 2016. But these statistics employ the public performance measure (PPM) methodology, which counts long-distance trains arriving at their terminating station within ten minutes of the scheduled time as on time.  Virgin Trains has recently began e-publishing data using the tight-time performance measure, where a train is only taken as on-time if it arrives at its terminating station early or within 59 seconds of schedule. By this measure, Virgin Trains achieved only 55 per cent punctuality over the period 1st June 2015 to 28th May 2016.

I think that it is fair to assume that the risk of service perturbation is greater on parts of the legacy network where capacity pressures are more severe, and the diagram below illustrates where the sections experiencing high capacity pressure (shown red) and medium capacity pressure (shown orange) are to be found on the West Coast Main Line (see footnote 4).

Source: Steer Davies Gleave

Source: Steer Davies Gleave

HS2 classic-compatible services will have the considerable advantage of by-passing the legacy network south of Milton Keynes, where path congestion is at its worse. However, there are red sections on the approaches to Manchester and Liverpool that will not be avoided. Furthermore, this by-pass will not necessarily protect classic-compatible trains from the knock-on of perturbations arising south of Milton Keynes. Also, by my count (see footnote 5) HS2 Phase 1 classic-compatible trains running north of the Handsacre junction on the West Coast Main Line (WCML) will increase the utilised train paths by a net three (see footnote 5), so potentially increasing the congestion on a section that is already rated as medium capacity.

The potential for legacy service perturbations to disrupt the smooth running of the HS2 network indicates the necessity for incorporating buffering into the timetable and supporting this, where appropriate, by the system design. Some consideration of the way in which this should be done is provided in Sections 5 and 8 of the SYSTRA report and we also  have the benefit of some comments by Professor McNaughton that are reported in an article in Rail Magazine from July 2012.

Aside from captive trains running both ways between London and Birmingham, which should be largely immune to legacy service perturbations, the least affected classic-compatible services on Phase 1 are likely to be those leaving Euston. Platform restrictions will require each platform to service two trains an hour (see footnote 6), dictating a target train turnaround time of 20 minutes. Aside from the, perhaps generally unlikely, possibility that delays in train servicing or discharging and loading passengers will cause a late departure, the main risk would appear to be the late arrival of the incoming train and even perhaps no platform being available immediately.

The SYSTRA report recommends that the headway used for timetabling northbound trains is relaxed from the minimum of 2mins 30secs to 3 minutes and that two of the resulting 20 train paths are kept spare, thus maintaining the planned 18 paths available for scheduled services. This, according to the report, “effectively creates a regular spacing of the delay margin” that “will make it possible to quickly recover small delays and return to a normal service” by the delayed train travelling faster than the timetabled speed (see footnote 7).

Of course, even if these recommended measures succeed in delivering classic-compatible trains to the Handsacre junction dead on time, then the trains will be directly vulnerable to any delays that the WCML may throw at them, and the designers of HS2 are powerless to take any steps to prevent this happening. Late arrivals at northern stations that are the result of WCML delays will I suggest, as far as the passengers are concerned, be a bad reflection on the HS2 service: passengers are unlikely to recognise the attribution of the delay being outside the control of the HS2 operator.

The SYSTRA report identifies the southbound direction as having “more potential for delays from trains coming from the classic network” (in Sub-section 5.5), but my examination of what can be done in that case will have to wait until part 3.

(To be concluded …)


  1. See Sub-section 8.1 of the report Operational Concept Study, Technical Note: HS2 Capacity and Reliability, SYSTRA, October 2011.
  2. See Sub-sections 5.3 and 8.1 of the SYSTRA report.
  3. See Sub-sections 8.2 and 8.3 of the SYSTRA report.
  4. The diagram is a section of Figure 5.1 in the technical report Capacity on North-South Main Lines, Steer Davies Gleave, October 2013.
  5. The HS2 Phase 1 indicative specification shows four conventional WCML services removed and seven HS2 classic-compatible services added to this line section.
  6. See paragraph 2.3.6 of the HS2 Ltd publication HS2 Project Specification, January 2012.
  7. See Sub-section 5.5 of the SYSTRA report. I mention in footnote 7 to part 1 that SYSTRA recommends timetabling on the basis of a maximum speed that is less than the achievable maximum “to allow late-running trains to speed up to recover lost time”.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: