A Layer Model for the Railway Infrastructure Design Process

Fixed-interval timetables are central to the further development of rail networks because they better meet customers' needs. However, infrastructure is needed to make the service possible. It would be helpful to be able to derive the track topology directly from rail service, but to do this, several open questions need to be answered. This thesis aims to find and name these open questions for the design of service-oriented infrastructure. Inspired by ideas and discussions about the cycle of track topology design, a working hypothesis for improved integrated transport planning for a more transparent railway system design was developed to justify service-oriented infrastructure. Three research questions regarding the relationship between rail service and track topology were formulated from the working hypothesis. The method of design science research was utilised to answer the research questions. Design science research is centred around an artifact that is constructed in cycles. The developed artifact is a railway layer model with several layers between the track topology and the rail service. The design science research method began with assembling an initial artifact and modelling one layer per cycle upon the initial artifact. The modelling was done with an example railway network and the validating functions of the current layer in a cycle. The modelling was undertaken in the direction of rail service by beginning at the track topology layer. Reversing the direction after finishing the artifact was carried out to chart the railway infrastructure design process for a service-oriented infrastructure. As a first result, the layer model consisted of collections of data with a specific purpose. These collections were used to find a suitable reverse path for the design process for a service-oriented infrastructure. The second result was clusters of collections named motifs. There are eight motifs in the landscape of problems for direction rail service to track topology. Some motifs can already be considered solved, while others still require further research.