Safe, high-quality and more frequent rail services are desired by both society and passengers. Since large scale infrastructure extensions to accommodate higher train frequencies are hardly practicable, a more intense network utilisation is considered the way to achieve these aims. As a result, trains will get in each other’s way more frequently, which can lead to an increase in delays and incidentally to safety hazards if a train fails to stop. In order to understand how all objectives can be achieved simultaneously, without the disadvantages of a more intense network utilisation, a model is required that takes both the objectives and the risks of a process into account. Such a model, the so-called safe envelope of operations, has been developed in this thesis. By taking a functional view, it allows to combine human, technical and organisational factors; three elements that each play a role in railway design and operation. An extended survey of railway practitioners and literature has been performed to see how the railways are controlled in practice and how this was initially intended. A main conclusion is that improvements to the railway’s real time controllability are necessary to solve conflicts between train movements in an anticipatory way. Also several instances were found in which design changes and informally developed practices led to surprising results in practice. Information about actual operations must be shared to avoid operational surprises. (Author)
Abstract