• Our booking engine at tickets.railforums.co.uk (powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Measurement of the impact of delays and disruptions to services

Status
Not open for further replies.

70014IronDuke

Established Member
Joined
13 Jun 2015
Messages
3,699
Is there any research into the impact of poor service levels on passenger usage? Or freight, for that matter?
If not, why not? Especially when we have universities offering degrees in transport and the like?

This site has daily reports on serious delays to trains and widespread cancellations. Every time the wind blows a bit worse than usual, there appears to be an awfully good chance of a pantograph snagging the wires somewhere between Peterborough and York.Or there is a fatality somewhere. Or a failure blocking the line with trains waiting hours for relief.

And whenever a suggestion that spare crews/capacity/operating staff or whatever should be on call, the default answer comes that we can't afford to have folks/equipment hanging around for the odd times that they are needed.

Sounds plausible, but how do we know, unless we can estimate the damage to passenger and freight demand caused by disruptions?

Every time a new route opening is discussed, or a current route upgrade is proposed, any suggestion that the new or upgraded route could have value by dint of acting as a diversionary route (or increase the capacity of a current diversionary route) – is dismissed by contributors (who it would appear to have worked in planning) that such considerations do not figure in the calculations of those who evaluate new routes or upgrading of current alternative routes.

But again, how can anyone say this unless we know the hit on demand caused by disruptions?

IIRC, in the early days of Inter-City travel, BR reckoned that 1 mph increase in average speed typically resulted in a 1% increase in passenger demand.(If I've got that wrong, I'm sure someone will quickly correct me.)

Of course, I don't know how provable this was, but surely, in these days of vastly improved train performance monitoring and ticketing sales data, some estimates could be made?

Obviously, to some extent the elasticity of demand will depend on the local circumstances. Long-suffering Southern commuters would appear to be almost impervious to a disastrous performance, the vast majority simply have to travel by train – whereas, say, a sustained poor performance of the Heart of Wales or Cumbrian coast services would probably result in a far sharper (relatively) drop off in demand as people turn to use or buy cars, or use buses or taxis.

But what evidence is there to show this, and map it out mathematically?

We certainly seem to have a vast supply of potential case studies. So, what is the state of research on this vitally important aspect of forecasting demand trends? Anyone know?
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

coppercapped

Established Member
Joined
13 Sep 2015
Messages
3,099
Location
Reading
This doesn't quite answer your question in detail, but it is a good background to using the rail passenger forecasting model.

There is also a paper by Paulley et al from 2005 reporting on the demand for public transport and one of the parameters considered was quality of service (QoS). The factors they treat under QoS are not the same as those you list, but I see no reason why considerations of delay, cancellation and overcrowding should not also be considered under the same heading. This report may be found here.

Added in edit: Obviously some academic work has been carried out in this area as can be seen by the abstract to work done by Starita and Scaparra Passenger railway network protection: a model with variable post-disruption demand service and published in the Journal of the Operational Research Society in 2017. The reference may be found here, but the full paper costs £35!
 
Last edited:

70014IronDuke

Established Member
Joined
13 Jun 2015
Messages
3,699
Thanks for this. I have had a scan, but need more time to read properly and assess.
 

Dr Hoo

Established Member
Joined
10 Nov 2015
Messages
3,976
Location
Hope Valley
The well-established Schedule 8 performance regimes are, of course, calibrated in an extremely detailed fashion on the best available 'marginal revenue effect' of cancellations and delays. These rates are derived from current versions of the Passenger Demand Forecasting Handbook, originally established under British Rail. Monitoring points are weighted by number and type of passengers, etc.

(This does not include 'value of time' to the individuals affected or immediate 'costs' such as refunds and taxis.)
 

Bald Rick

Veteran Member
Joined
28 Sep 2010
Messages
29,211
Yes research and analysis has been done by the TOCs. It's not public. The travelling public are rather resistant to delays - clearly there is much complaint, but generally they still travel. A TOC MD (of a London commuter operwtion) once told me that PPM would have to be consistently below 80% for months for there to be any discernible affect on revenue. Interestingly GTR have had a good go at this.

Nevertheless, the OPs question is answered by the industry in a different way. As Dr Hoo says the contractual incentives in the track access contracts are calculated in a way that reflects revenue lost. In the event that Network Rail causes poor performance over a sustained period, there is a mechanism for increasing the compensation called (surprisingly) sustained poo performance. On top of this, franchised TOCs will have penalties to pay the franchise specifier if train performance falls below certain benchmarks, although the measures vary by TOC.

These contractual incentives enable the operators and Network Rail to justify investment in performance improvement initiatives. Operators will provide thunderbird locos where the cost of the loco and crew on standby will more than pay for the reduction in penalties to NR and DfT for TOC caused incidents (usually train failure).

The converse is that as NR are responsible the larger portion of delays, TOCs are (perversely) incentivised to save cash on things that help to ameliorate delays in general regardless of cause, e.g. the provision of spare crews. They save the cost of the crew, and as reactionary delays increase, the TOCs get more in performance payments from NR (on average).
 

SamYeager

Member
Joined
20 Mar 2014
Messages
339
In the event that Network Rail causes poor performance over a sustained period, there is a mechanism for increasing the compensation called (surprisingly) sustained poo performance.

I'm sure it's just a typo but it did make me smile. :)
 
Status
Not open for further replies.

Top