Much has been written on the deployment of the European Railway Traffic Management System (ERTMS) Level 2 across Europe. Most of the more recent High Speed lines are equipped with the system (or more accurately the ETCS – European Train Control System plus GSM-R) where infrastructure and trains are more or less self contained and the supply base has been a single contractor.
Those railways that have begun to roll out the system on existing lines have encountered many problems with performance, interoperability between different suppliers’ equipment, retro fitting of rolling stock and interfacing the system to legacy signalling equipment.
In Denmark, 60% of the signalling equipment will have reached its end of life within 15 years and its ATC system equipment (more correctly Automatic Train Protection) will be unobtainable after 2020. Typically the signalling equipment is 50 years old, with 40% of interlockings requiring renewal by 2015. 50% of train delays are due to faults in the signalling system.
Faced with this situation, the rail authorities have decided to implement ERTMS nationwide, ridding itself of all existing signalling equipment in the process and retro fitting rolling stock other than that with only a short remaining life.
So, how was this decision reached and how will the system be implemented?
Justification and Procurement
Like most European countries, the Danish Rail organisation is split. Banedanmark, Rail Net Denmark, is responsible for the maintenance of 2,323 kilometres of track and the traffic control and signalling systems. The principal main line train operator is DSB (Danske Statsbane – Danish State Railways).
Recognising the growing problems with signalling obsolescence and reliability, in 2006 a strategic political decision was taken by the Danish Parliament that a complete renewal programme should be embarked on over a 10 year period subject to suitable business and technical projections being prepared and approved.
This analysis took place between 2007 and 2008 and needed to prove that such a programme would benefit from the economies of scale. The new system would have to offer a quantum leap in technology while using standard designs and components, resulting in operational savings and a reduction in maintenance costs. The adoption of simple and safe operating rules should reduce the number of safety approvals needed and result in fewer traffic management sites.
A complication arose with the Copenhagen S-Bane, this being a regional suburban service for the city sometimes adjacent to the main lines but never sharing tracks as the electrification system is different. It is broadly equivalent to the London Overground in its role. This would not only need signalling renewal but also a significant increase in capacity. As such, a different solution for the S-Bane has been adopted.
Once agreed, a procurement process proceeded with many firms competing for the considerable business on offer. Eventually the main contracts were awarded in 2012 on the following basis:
Two ETCS contracts; East and West, the rationale being that no single company would be capable of supplying the entire country (Alstom has the East sector and a Thales – Balfour Beatty Rail consortium the West sector);
» One contract for the train equipment on both main line and regional rolling stock; » A supplementary GSM-R contract to include GPRS packet switching capability and additional coverage to the GSM-R voice network already being rolled out;
» An ongoing framework contract to provide for equipping any new lines;
» An option for ERTMS Level 3 (Regional ERTMS) on rural lines.
Taking this approach has resulted in considerable cost reductions compared with the European norms for ERTMS provision both trackside and train-borne. A separate contract has been let for the modernisation of the S-Bane.
Contractual Collaboration and New Rules
A project of this size needed some new thinking if the all too often contractual barriers were not to be encountered. During the tendering phase, day long sessions with all involved parties were held to discuss and understand both technical and non-technical issues. The need for close collaboration was self evident but identifying the potential problem areas at an early stage was essential. In this way, the suppliers got a much better understanding on what was required and the submitted bids were thus of a higher than normal quality.
The sharing of risk is always contentious and both client and supplier had to accept a pragmatic approach. One factor within the collaboration process has been ‘how do we disagree with each other?’ Or, in other words, when disputes occur, what is the mechanism for resolving them? As someone said, ‘if the lawyers appear, then we will have failed’.
A dominant element has been to ensure the resulting system and equipment is as near to a standard design as possible, with any necessary developments and deviations having to be proven as absolutely essential. This caused a considerable focussing of minds and resulted in both Banedanmark and the train operators needing to change their organisations to suit the new signalling system. These changes will affect controllers, dispatchers and maintenance staff, as well as drivers and other train crew.
More contentious has been the decision to modify the operating rules to fit ERTMS deployment and remove some long cherished national principles, known locally as ‘Kill your own Darlings’. The outsourcing of project management was also rejected on the basis that integration would suffer; all parties have been required to participate in this activity.
The magnitude of the project makes the timescale for implementation quite long. Since the Danes have had no previous experience with ERTMS, the start of the roll out is to be two trial lines, one in the north between Langå and Frederikshavn, the other nearer to Copenhagen between Roskilde and Næstved. These will be commissioned in 2016 and will give both infrastructure contractors vital experience for the ongoing programme.
Banedanmark will take its time to learn from these early deployment routes so that further commissionings can be sure that all technical and operational situations have been understood and any problems rectified. The main lines will follow between 2018 and 2021 with the rural lines following on from this. It is possible that the latter will adopt the Regional ERTMS system currently being both trialled and implemented on similar lines in neighbouring Sweden.
The total cost of the project is budgeted at 14.8Bn Danish Kroner equivalent to around £1.5Bn. Some 20% of this expenditure will be by central government grant.
The installed system is expected to have a design life of 25 years and both Banedanmark and its contractors are expected to continue in partnership for this period. It is accepted that some updates of equipment and components will need to happen during this time but these should be in line with what happens to other ERTMS systems across Europe.
Lessons for others
It would be easy to suggest that if the Danes can do it, why cannot the UK and other countries within Europe do likewise?
The situation is not that simple. For a start there are the logistics of scale; the Danish network is 2100 km with about 1600 trains per day (excluding the S-Bane), so the network is much smaller. Equally, past investment in signalling systems has been proportionally less than in other countries so it has a much older age profile. The more modern systems can afford to be retired early.
Even noting this, the Danes concluded that the supply industry could not cope with nationwide roll out using a single contractor. The UK ERTMS programme for main line deployment firstly on the GW and then the ECML is probably as much as both the client and supply industry can cope with in the timescales declared.
That said, full marks to the Danes for being brave enough to take this bold step and for putting a package of collaboration together that should overcome the perils of interoperability, contractual disputes and rule book changes. We shall all watch with interest the next few years to see if it pans out as planned.
Copenhagen S-Bane Control System
The S-Bane connects the outer suburbs of Copenhagen with the city centre. It too requires a control system update, not only to replace outmoded signalling equipment but also to provide additional capacity.
The rail authorities looked at options and concluded that to incorporate the S-Bane network into the main line ERTMS programme would not be a good idea for three reasons. Firstly, the ability of ERTMS to offer the traffic throughput required was suspect, secondly, the capacity of ERTMS in heavy traffic areas such as city centres was known to be limited owing to lack of bandwidth in the circuit switched GSM-R configuration and thirdly, the timescale for completing the S-Bane upgrade was unlikely to be commensurate with the main line programme.
Being as the network is separate from the main lines, it was feasible to adopt a different solution. The advent of Communication Based Train Control (CBTC) systems elsewhere in Europe led to the easy decision of adopting one of these. After competitive tendering, a contract was awarded to Siemens for its TrainGuard MT wireless based system.
As well as the CBTC element, the contract includes signalling interlockings, traffic management, datacomms including passenger information, facilities to provide extra functionality and train equipment associated with the signalling system. As for the main line, an important factor has been that the system is well proven and in service elsewhere thus hopefully ensuring no nasty surprises with either the technology or operation.
An early deployment line from Jægersborg to Hillerød will be equipped first to test out the system and enable training of staff to take place. Thereafter the policy will be to commission sections from the extremities of the various lines and work inwards towards the city centre. The central sections will be the last to be done. Once a section of line is commissioned, the changeover at the nominated point from existing to new will be by a simple switch on the train and thus trains have to be equipped with both old and new technologies.
Altogether, there will be 6 phases and the whole project will be implemented between 2014 and 2018 at a cost of 3.9 billion Danish Kroner, equivalent to around £400 million.