“Anything that sits outside of that, we need to be exploring different models. You first of all need to look at the nature of the investment that you want to do. So at one end of the spectrum you can have Digital Railway, or you can come back round and say ‘let’s look at the Brent Cross station development on the Midland Main Line’. You then have some of the things we’re looking to do with Leeds City Council in terms of station development, and then you have things like the York Central development.
McIntosh reckons that each will take a different way of working. A property developer might be happy with the risks of investing in Leeds station’s commercial side, but wouldn’t take the risk with anything that might shut the railway. That risk best lies with NR.
“As long as NR is an extended arm of the Government, then that should be underwritten by the Government via NR. But you need to explore and identify the opportunities and look at the best model for doing those,” he argues.
But that puts us back to risk of delivery, and that’s where NR’s track record isn’t good.
“Agreed,” he replies.
So what’s the answer? How does NR change that?
“I think that some people would put their money with NR. But it’s about how you have the relationship to look at each individual opportunity, because what somebody might be willing to invest in Digital Railway might be massively different to what we might want to do around Leeds. But we need to move away from NR’s delivery record and start to explore different delivery mechanisms.
“Because you’re right, NR recently has had a difficult delivery record. But up until (I guess) the problems we encountered last year we’d been doing very well. And I was part of projects organisation that was delivering those kind of commitments. Jobs like Borders, and before that Airdrie-Bathgate and some of the station remodellings like Nottingham - these things have gone well.”
It’s worth pointing out that while it’s true that those three projects went well, two of them were largely away from live railways, which made them considerably easier. Just a few miles south of York lies Doncaster, another railway town. NR is building a new platform (Platform 0), linked to the main station by a new footbridge. Passengers should have been using it last December, but it’s not finished and is likely to be a year late.
McIntosh concedes: “Doncaster is late for exactly the same reasons as things like the Great Western Electrification Programme being late, in that the discipline around the early gateways of projects wasn’t adhered to. So Doncaster was allowed to go through its GRIP3 when it wasn’t suitable to do that. Only when you start applying the rigour and discipline to those early gateways will you begin to recover the position.”
For Doncaster, the problems centred on associated changes to signalling, particularly a plan to introduce bi-directional signalling south of the station. McIntosh notes: “The reality with Doncaster is that when they went through GRIP3, no one had properly understood the significance of the signalling changes associated with Doncaster, and the complexity of the work you’d have to do in the Westpacs and the relay room.”
He explains the wider problem with another example. “At the end of my first week as Route Managing Director, January 9 this year, we had a GRIP3 gateway for the King’s Cross remodelling, which is the biggest intervention we’re going to do on the LNE route probably for a generation. The team presented to me what they thought was a GRIP3 that was ready to go - they were saying we need to go through this gateway, we need to hit 2019 and it needs to be done.
“And very quickly I concluded that it wasn’t in a fit state to go through a GRIP3 gateway. So I didn’t allow it. The impact of that is we’ve delayed that project for a year. But it’s the right thing to do, because it’s that lack of discipline and difficult decisions at that point in a project’s lifecycle that allows things like Doncaster to happen.”
It is the checks and balances of project gateways that has been absent in previous NR schemes, if McIntosh is right. He’s keen that projects on his patch are thoroughly checked by those with the experience to find the problems, which contrasts with the view of Francis Paonessa, who heads NR’s infrastructure projects division. Speaking in February 2015 after major engineering overruns at Paddington and King’s Cross over Christmas 2014, Paonessa said that it wouldn’t have made any difference if more people had checked the plans.
NR used a technique known as quantified scheduled risk assessment (QSRA), which takes the results of previous projects to estimate the probability of a future project running as planned. Looking back over King’s Cross, Paonessa said: “We retrospectively went back and looked at the plan that we had, and even if we attended the QSRA with a hundred people, I personally would have been quite confident going into the plan that we had.”
McIntosh explains NR’s project problems: “It’s a lack of assurance. I’ve spent pretty much most of my career delivering projects, and as a project person you’re incentivised to deliver your project. What that does is sometimes creates a mindset where you push too hard for delivering a project and meeting your milestones. To counter that you need some assurance processes around your gateways to make sure that before your project moves from one state to another state, which in NR speak is GRIP, it’s fit and ready to do so. That’s where we’ve not been robust enough.”
His message is that there have not been enough checks to counter the natural optimism of project managers. He continues: “That’s no different to projects like Great Western electrification, which has the same issue on a grander scale. When Mark Carne went to the Public Accounts Committee, he talked quite openly that the project went into build status before its design was finished.”
Surely it’s obvious that you can’t build something before it’s designed?
“You convince yourself that you can build bits of it. That’s what happens. That’s why you need an assurance.”
