Dr Rudolf Schilder, head of ÖBB Infrastructure's line and station management business unit, explains how this has been achieved.

ÖBB operates a mixed traffic network of almost 10,000km which requires high track accuracy for passenger trains but faces rapid track deterioration from dense freight traffic. ÖBB Infrastructure is under constant pressure to provide optimum track availability at minimum cost without a deterioration in track quality. Key to this is continuous research, economic analysis, and implementation of innovative track maintenance and track components. ÖBB's strategy is based on life-cycle cost (LCC) analysis, with a comprehensive database now being used to monitor track condition, which allows a prognosis of track and track component deterioration based on long-term observation.

A number of information tools have been developed. For example, the New Austrian Track Analysing System (Natas), which integrates ground penetration radar measurement into the track report, is used to analyse track condition. Another important tool is a geotechnical database, where each 200m section of the entire mainline network is classified according to the condition of ballast, substructure, drainage and adjacent structures which allows optimal planning of rehabilitation projects.

Detailed investigation into the cause of track failures led to an improvement in track components, such as rails, fastenings and rail pads, which increases their working life and reduces the occurrence of track defects. The measurement and analysis of rail inclination is another helpful tool developed to indicate pad or fastening failure.

The correct elasticity of the whole track system is of major importance for a cost-effective track strategy, so ÖBB now installs pads under concrete sleepers and pays close attention to the condition of the ballast and formation. Rehabilitation of the formation always includes the application of a gravel-sand layer above a geo-synthetic layer.

Another new strategy is "integrated maintenance" where tamping and rail grinding is carried out in one common track occupation. This has enabled ÖBB to achieve an average output of 3.5km per shift and an internal rate of return of slightly more than 5%, even allowing for the additional grinding cost. Integrated maintenance also results in longer lasting track quality and fewer rail defects.

Traditional levelling-lining-tamping only delays the onset of long-term track deterioration, and track misalignment generally occurs again at the same location. A solution to this is design tamping where the idea is to over-lift the failure to delay its onset, which can be achieved without incurring additional costs.

For trackwork, ÖBB uses high-capacity machines under three to five-year contracts including:

• track-geometry maintenance machine groups with four-sleeper tampers, stabilisers and ballast distribution systems

• continuous-action turnout tampers with integrated ballast distribution

• ballast cleaning machines

• formation rehabilitation systems combined with drainage maintenance systems

• combined track relaying and ballast cleaning systems, and

• systems for transporting and laying preassembled plug-in turnouts.

While engineers must design infrastructure appropriate to commercial needs, the infrastructure must also be sustainable, because its viability will be affected by over or under investment. The only way to meet these disparate objectives is to develop a better understanding of the infrastructure and a deeper knowledge of the correlation of the various track components.

Key to this is continual research, analyses of economic and technical dependencies, and the implementation of research results in everyday maintenance work. More than 15 years ago, ÖBB, in cooperation with the Technical University of Graz, started LCC analysis which led to the development of a maintenance planning model. The principal maintenance activities were analysed and we started to create a database with significant data on track quality, maintenance work and basic track data, namely rail section, fastening type, sleeper type, installation date, and other related information. Later, information was added on specific locations, turnouts, level crossings, bridges, stations, platforms and geotechnics. As a result, billions of items of track data information from the rail network is now stored for use in statistical analysis as well as being available for normal on-site maintenance.

LCC has to address the entire product span, which means the whole service life of the railway system must be taken into account and not one part in isolation. Direct LCCs also need to be considered, namely the investment required, inspection, and maintenance, as well the costs of operational hindrances (COH).

ÖBB uses a linear cost model for evaluating COH based on average costs for standard trains. This includes provision for train delays, increased energy consumption due to braking and acceleration, additional train-kilometres caused by diversions, and the cost of railway replacement bus services.

Follow-up delays, train failures, difficulties connecting with other services, shunting problems, fee-repayments, and negative market responses are excluded. As a result, COH associated with track work planning, optimisation of track possessions, and evaluation of infrastructure availability must be an integral part of any track strategy.

Track behaviour is influenced from above the top of the rail by the type of rolling stock, traffic volume, and speed, and from below top of rail by the alignment, rails, sleepers, ballast and formation (pictured above). But which track components limit service life?

Rail wear depends on traffic volume, radius, profile, and steel grade, but rails can be changed easily so they do not limit service life. Hardwood sleepers can wear out but can be changed to a certain degree, whereas the service life of concrete sleepers is not critical, so again there is no limitation on service life.

On the other hand, ballast replacement or cleaning is sometimes necessary but expensive, so ballast has a strong influence on track behaviour and is therefore a critical element, as is the formation subsoil quality and drainage. If the latter is of insufficient quality, track optimisation is impossible.

As a result of our evaluation, the service life of tracks using granite ballast instead of limestone ballast can be extended by 20%, and by up to 35% for turnouts.

Depreciation is the key factor in the total annual cost of a railway. Longevity strategies aim to extend track life, whereas reducing maintenance and accepting a reduction of service life is highly uneconomic, and operating costs caused by maintenance work (or a lack of maintenance) are significant.

The main driver of cost optimisation is an extension of service life to reduce depreciation costs. A reduction in COH is possible by changing strategy or working duration, or by running fewer trains, which is of course the least desirable option. Maintenance costs can be cut by using better quality components and optimising the maintenance regime. Having a higher initial quality by using more expensive track components is economic provided it results in a longer service life.

It is worth noting that good track behaves well and lasts longer between maintenance interventions, whereas poor track deteriorates more quickly.

Since 2000, ÖBB has compiled and stored more than 3 million e-functions with all relevant information and history for the core network. This database gives us a functional knowledge of track quality behaviour which enables us to predict quality behaviour thereby optimising the time at which track work is undertaken, which in turn increases track life.

Last year we launched our Expert System for Infrastructure Facilities which combines many different tools, and provides the regional track engineer with access to a custom-made proposal for his facilities. The system comprises a precise record of the network's status, a prediction of deterioration, a list of recommendations, and an LCC evaluation of maintenance or renewal. For the first time it makes it possible for the track maintainer to assess track works in terms of quality, transparency and objectivity, with all data stored in a database accessible to all technical staff and regional engineers.

Natas was introduced in 2003 to aid maintenance planning. All the data collected from the track recording car is linked with other information and then edited graphically, which makes it possible to predict infrastructure deterioration. Natas normally provides an overview of 5km sections, but it can also provide general remarks for 50km sections as well as details displayed in special windows to aid analysis of deterioration over 100m segments.

Our comprehensive databases built up over several years for the entire network now enables us to monitor track condition and thereby produce a prognosis of track and track component deterioration. This, combined with the introduction of tools such as Natas, new methods of tamping, and the use of high-capacity machines, has helped ÖBB to simultaneously improve track quality and reduce costs.