Project Comparison: Antwerp Centraal vs Stuttgart 21

25 Aug

Over the weekend I posted an article about the controversial Stuttgart 21 project (you can read it here). The idea of that post was to provide an introductory guide to the project and the associated protests, along with a commentary on the aims, goals and points of contention.

This post, in way of comparison, will take a look at another major European station that underwent a similar rebuild project, Antwerp Central, and draw conclusions by comparing the two projects.


Antwerp Central station shares many aspects of form and function with Stuttgart Hauptbahnhof (Hbf). Both are regarded as great examples of architecture and design, both stations serve a large city and both are transport hubs for local, regional and international rail services. Up until March 2007 both were also terminus type stations, but since then, due to the construction of a tunnel, trains can now call at Antwerp Central and continue their journey without having to reverse. Similar to the Stuttgart 21 project again is that the construction of the new parts of the station took place below the old tracks and not as a new development alongside it or elsewhere. This is how it looks today (my own picture);

The trains arrive and depart from three different levels, with the top two levels being divided into 2 parts either side of the atrium. You will see a train on the top right just above the grey concrete and the four rows of white lights indicate each part of the lower levels -1, and -2.

You will also notice the striking architecture of the old station together with the glass and steel train shed both of which were built in 1905 and how they are still a major part of the complex today. As preservation of the original building is a key issue in the Stuttgart 21 project, it is interesting to note how the original architecture has been largely kept intact with this vast modernisation and how the issue of people moving around the inside of the structure has been well thought out and planned.

Efficiency Gains Through Modernisation

  • Through trains no longer have to reverse, reducing what is known as “Pathing Conflicts”, when trains have to cross a number of tracks either entering or leaving the station which prevent other trains moving in the other direction.
  • Passengers looking for a connection don’t have to walk all the way down to the end of the station, then back again to their next train, by being able to use the 3rd dimension of height the designers have reduced connection access times.
  • Calling trains also save time on their journey as the driver doesn’t have to switch ends of the train so stopping times have been reduced.
  • Less real estate had to be purchased for the development due to the underground nature of the project.
  • With 14 platforms the station is now able to handle more trains and the design is considerably more future proof. Capacity was a big issue for Antwerp before the modernisation.

All the above points are directly comparable to those on the Stuttgart 21 project.

Project Costs – Are We Comparing Apples with Apples?

To compare the cost of the two projects let’s outline the scale. To do this I’ll use a rudimentary breakdown of the Antwerp project costs.


  • €765m = Total project cost, as is documented here on the Railway Technology website.
  • €306m = 40% of the project spend was for the station itself (again this figure is documented on the above link). From this figure we can estimate that…
  • €306m = 40% would have been spent on the 3.2km tunnel that now links the North and South of the station. This in turn leaves…
  • €153m = 20% for expenditure which can’t be attributed directly to either station or tunnel including new track that isn’t in the station or tunnel, financing costs, legal bills, inflation, project add-ons, unforeseen expenditure etc;

From this we can calculate a cost of €90.5m per tunnelled kilometre.

Backing this figure up are the project costs for the Gotthard Base Tunnel and the Lötschberg Base Tunnel in Switzerland both of which are giant transit tunnel projects. These figures come out at;

  • Gotthard = €7.2bn for 57km tunnelled giving an average of €126m per tunnelled km.
  • Lötschberg = €3.5bn for 34.6km tunnelled given an average of €101m per tunnelled km. (This figure includes the estimated €1bn extra to complete the 2nd bore as referenced here.)

Taking an average of these 3 figures and allowing for inflation, and the fact that the two Swiss tunnels are going through mountain rock, whereas Antwerp’s tunnel goes only under a city with softer geology, we can envisage average building costs coming in at around €110m
per tunnelled km.


Now today, from the official Stuttgart 21 website there is this data; (Which incidentally does not cover the separately budgeted high speed rail link to Ulm.)

  • 33km of new tunnel and cuttings (A “cutting” is a railway in a shallow gorge type of environment). Having looked at the map 85% of the new route will be tunnelled.
  • 3 new stations: The new Stuttgart Hbf (where the railway will also be reorientated by 90 degrees), the station at the airport and a new S-Bahn local station at Mittnachtstrasse.
  • The tunnels and lines will be built for speeds of up to 250km/h, which Antwerp’s tunnel is not.
  • The current estimated budget is €4.1bn

So let’s see how that cost estimation stacks up;

  • 85% of 33km (the amount which will be tunnelled) is 28.05km. Multiply that by the average cost of tunnelling as discussed above and you get €3.085bn
  • The remaining 15%, roughly 5km, will be open track. The Cologne to Frankfurt high speed line has a cost (According to Deutsche Bahn figures) of €33m per km. Therefore using this ratio will result in a cost of €170m.
  • Stuttgart Hbf. Taking the cost of Antwerp Central, and adding the whole station will be twice the size of Antwerp’s underground sections, we can estimate a figure of around €950m
  • Stuttgart Airport’s new station will also be underground; we have Frankfurt Airport station as a precedent, which cost €225m 13 years ago. It is logical to assume the cost will be roughly the same as Antwerp’s’ or the cost of Frankfurt Airport’s in today’s money, therefore €306m
  • Stuttgarts new local transport station at Mittnachtstrasse will be a more modest size and on ground level so we can assume a lower ratio/cost here. I estimate €150m
  • The new depot and sidings for train maintenance at Unterturkheim will also be to a more modest figure, again let’s say, conservatively (!) €150m
  • This gives us a sub total of €4.811bn
  • When we add on the 20% for other project costs, unforeseen issues and overruns that was attributed in the breakdown of Antwerp’s costs we have to revise the total project cost to €5.7bn

This, as you can see, is now considerably more than the €4.1bn figure currently being advertised as the cost of the project.


The Stuttgart 21’s objectives are sound. Building the new rail link will remove bottlenecks in the regions transport infrastructure. It will also remove “Pathing Conflicts”, increase capacity and improve connections, much in the same way as is now being enjoyed at Antwerp Central. The effects of a speedier journey will also be felt much further afield and linking the airport with the city promotes integrated transport which is a must for today’s mobile society.

The costs however are a major worry. Even with my rudimentary calculations based on ratios and evidence of breakdowns from previous projects I can demonstrate that the current budget for the Stuttgart 21 project is already under pressure.

It is not clear to me, nor is it clearly reported, what the budgets are for each sub-section of the project. A more open approach from the project sponsors to the general public would be a good PR move, as currently they seem a guarded secret. But then perhaps this reflects the political pressure that the budget is already under. Rising costs are giving the protests a louder and more credible voice. It is time for the project stakeholders to demonstrate how their figures have been calculated, using credible sources and precedence, and then open them up for presentation to the public. Without this, Stuttgart 21 will remain under pressure from the public and its own demonstrably tight budget constraints with the result that a question mark will continue to hang over the project’s future.

Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

One Response to “Project Comparison: Antwerp Centraal vs Stuttgart 21”

  1. Gunter Sharp November 29, 2010 at 12:30 am #

    Thank you for such an informative article. I just learned about this project after reading a Stern article about the mediator, despite having visited Ludwigsburg, a Stuttgart suburb, three times during the last five years. Your approach to cost analysis appears to be more realistic than any other that I have seen in my professional experience. If I teach another course in engineering economy (I’m semi-retired), I will certainly use Stuttgart 21 as a case study.

Leave a Reply