About railML
railML® is an open-source railway markup language designed to facilitate seamless communication among diverse railway applications. The data structure of the railML® interface is based on the XML language. It is described using the widely supported and standardised XML schema definition language (XSD) which allows syntactic validation and generation of parser code.
A wide range of extensive computer programmes exist for railroad planning processes. In today's world, connecting various railway software packages often presents significant challenges. railML® offers an XML-based solution for simplifying data exchange between railway applications.
Beside the machine-readable description of the standard, railML® offers comprehensive documentation to support the efficient implementation of interfaces and easy understanding as well as an active community that can be used to quickly resolve any remaining ambiguities.
Image: Comparison without and with railML®
The image compares two network setups: one “without railML®” showing 10 interfaces due to direct connections between all nodes and second “with railML®” showing only 5 interfaces with a central hub simplifying the connections.
Background
The railML.org initiative was founded in early 2002 in response to the background of the constant challenges of connecting different railway IT applications. The Fraunhofer IVI Dresden and ETH Zurich laid the foundation for the development of the railML® interface, which meets this demand for standardisation using XML.
The first railML® schema was introduced in 2005, followed by railML® 2.0 in 2009. Between 2014 and 2017, the close cooperation with the UIC brought in state-owned railway partners (ÖBB, DB, SBB, SNCF). After the UIC project ended, cooperation with railways intensified, which enabled railML.org e.V. to advance the development of railML® 3 version and continue the maintenance of the RailTopoModel for the railway sector.
Community
The development of railML takes place in working groups of experts in the area of railway software, data modelling and IT experts. The working groups are defined by the five different subschemas of railML. Each working group meets separately once every four to six weeks to discuss content issues and further development steps. The results of railML.org are available for download after agreeing to our licensing conditions. You will find XML schemas, sample data files and the detailed railML specifications. All interested institutions and businesses are invited to contact railML.org and participate in the railML.org initiative.
Additionally, railML.org hosts biannual conferences, uniting the railML community for vital discussions and collaborative efforts. railML.org partners share their railML applications, project updates based on railML. The spring conference predominantly takes place online, while the autumn conference is facilitated by our stakeholders in person, catering to the global railML community. These events are typically conducted in English and if requested by the community also in German. For further information about the dates of our upcoming conferences and meetings, see our event schedule.
Open data
The 5-Star-Theory of Tim Berners-Lee defines 5 levels for open data, as well as their costs and benefits for users and publishers. According to this theory, the railML schemas follow three stars, meaning they are open non-proprietary data. This availability allows users to manipulate the data freely, without the necessity of owning any proprietary software package.
International industry standard
Since 2022, railML has been established as an ISO standard (ISO/TS 4398) worldwide under the marketing name RailDax (Railway Data Exchange). The purpose of RailDax is to facilitate the planning of railway operations between organisations in the transportation sector and enable efficient, more accurate and less ambiguous data exchange.
RailDax has been developed on the basis of railML 2.5 (railway Markup Language 2.5), which is managed by railML.org. It provides an open XML-based data format that allows data to be exchanged between different applications: connecting information on the infrastructure, rolling stock and timetable data required for capacity management and timetable planning. The standard is intended for railway and transport authorities, infrastructure managers, train operators and rolling stock manufacturers.
Currently, railML.org, with the support of its partner Jernbanedirektorat is planning to update the standard to railML 3.4 between 2025 and 2027.