Thread by @garius

Nope. Crossrail's problems aren't really budget related. They're hubris (which led to slack project management) and the sheer complexity of modern signalling.

The last bit is important, and I think people (even in the industry) don't full get that yet. Let's talk systems /1

https://twitter.com/tyteen4a03/status/1073531276446101504

So when people think of signalling on the railway, they think of it as being comprised of two distinct parts: the stuff which says what can move, when and the train which in some way acts upon that instruction /2

What those two things look like has changed drastically over the history of the railway, but the RELATIONSHIP has always been the same. Something at track side decides if a train can move and then it does - whether lamps and steam trains, or electric pulses to electric trains /3

What's happening now though, with this generation of signalling and rolling stock, is the first genuine EVOLUTION in that relationship for perhaps 50 years.

To oversimplify a bit, on the Elizabeth line the decisions will happen ON THE TRAIN as much as at track side. /4

Now Lizzy ISN'T unique in this. It's a quasi-early adopter, but this stuff is happening elsewhere and everywhere will shift to it eventually. But what it does mean is that a lot of the ways we build trains AND signalling systems are changing /5

Because signalling manufacturers and train manufacturers used to (largely) be able to agree a set of specs then go their separate ways, ignore each other and do their own thing. A nice, clear and (relatively) simple demarcation of responsibilities.

Not any more though /6

Now the signalling manufacturer has to build some bits trackside, but some stuff into the train software too. And the train manufacturer has to build the train and software to run it, but also make that software CAPABLE of being tweaked and augmented by the signalling packages /7

And as both Siemens and Bombardier are discovering here (as are they, and others, elsewhere, e.g. in Hong Kong) is that this is an UTTER game-changer in how you have to cooperate to design, build and integrate railways.

Because none of the old rules of working apply /8

As both Siemens and Bombardier are discovering, this isn't Kansas anymore, Dorothy:

Railways aren't just one system integration project any more kids. They're TWO - trackside AND trainside. /9

And that's a key reasons why doing the trains on Lizzy is proving tricky, which the problems with the build EMPHATICALLY haven't helped. Because EVERYTHING is taking longer to test and perfect, not because anybody is doing a bad job, but because this shit is:

1) New
2) Hard

/10

Who gets to push software? How are bugs prioritised? How do you identify which layer (signalling or ops) that bug lives in? Who is responsible for doing/paying if a code decision in one layer is causing a code issue in another?

All these issues have to be picked through. /11

Which will all happen. Siemens and Bombardier need it to happen, and Howard Smith (Lizzy's DOO) is very good at finding a path through this shit from the operator side.

But it is hard. It is new. It is taking time.

And it's not a problem you can throw money at to solve. /12

So a lack of money is a SYMPTOM of Crossrail's problems but it is largely not the cause. What this is probably just showing is that the 2010 rescoping was what a lot of us said it was at the time: bollocks

The project's going to cost what everyone said it would to begin with /13

Now before anyone says it yes - there are wider signalling issues than this (the number of systems involved, problems at Heathrow etc) but I just wanted to highlight what was different HERE from problems people would recognise already and we'll cover those in a future article /14

Hope that's useful though!

Remember, you can read our full breakdown of Crossrail's problems here: londonreconnections.com/2018/crossrail…

Back us on Patreon here: patreon.com/lonrec

And, as it's Xmas, if you want to buy me a coffee you can do so here! ko-fi.com/garius /END

ADDENDUM: Based on my mentions for this and over the last few days, I figured it might be quite useful to create a handy flowchart for those wanting to understand how Crossrail's problems relate to HS2. @rail