Unfavourable rail network numbers
// April 29th, 2008 // Comments // travel
I have come to the conclusion, after being without a car for five years, that travel by rail in the UK is just about the most infuriating process you can undertake on a regular basis. Every day, one of my 6 train journeys to or from work will fail in some way or other. Either there will be a delay, a last-minute platform change, a cancellation… something goes wrong.
Looking at the complexity of the rail network, it’s not surprising that it doesn’t always run exactly right. However, since I can pretty much guarantee that my train journey to work or back will be compromised in some way, I’m starting to wonder. There is no ‘normal’ service, you see. It’s a myth that the trains usually make it on time. Perhaps it is more than 50% of each leg of any given journey… maybe. But at some stage in every multi-sectioned journey I’ve had, some aspect of the trip fails.
II’m beginning to wonder what could possibly fix this situation. I’m told daily by an electronic voice: “I’m extremely sorry for the severe delay to this service.” I wonder what rail executives hope to achieve by creating an automated apology system? It’s not exactly like it’s a good value for money trade-off either. Train travel is bloody expensive, and it doesn’t work the majority of the time.
Maybe executive heads need to roll, maybe a huge infusion of public money needs injecting into the system, maybe… they need a change of perspective. If they measure every individual step in a day, they might come to the conclusion that they’ve failed n number of times. n=number of delayed, cancelled or otherwise compromised phases of transit between any two given points on the network. However, if X represents the full journey of an individual in a sampled population, and X is made up of a number of phases, I wonder how many X’s would end up compromised in some way?
Sample 100 complete journeys across the entire network. Each journey is an individual X (complete from point A to B with each stop and change included). Each completed leg of any given X could be designated y, and any compromised legs (including additionally-accrued attrition from one failed leg compromising another) could be w’s.
There are two numbers I’d be interested in seeing:
- The ratio of y’s to w’s across the sampled population’s collective X’s
- The number of X’s which contain at least one w
I reckon and predict, that these numbers would reflect very unfavorably on the network’s ability to run itself.
What’s the fix?
No idea… though I do worry no one in the rail network has any clue either.



It’s the transition which hits you. From begging your interlocutor to understand you—to even bend a little in your direction—to being understood so effortlessly but still being in an unfamiliar place. This is traveling from France to Amsterdam.




