A resilience state model for a railway system is proposed consisting of three boundaries putting pressure on the operating state: Safety, Performance (Capacity & Punctuality) and Workload. In order to model the pressure of the boundaries, an additional dimension is added where the slope represents the pressure. By doing so, the model is able to differentiate between internal changes that keep the system in a resilient state or have it move towards brittleness. The resilience state model is also used to develop a quantitative signal model, indicating pressure change of the boundaries. A newly defined resilience signal (RS), a quantitative indication of change in system resilience, can be created with help of the signal model and be used for anticipation during operations. The resulting parametric functions will be evaluated and tuned by empirical testing in further research. Using data from governmental reports on responses to incidents, two empirical cases are worked out using the signal model. The first case shows the correlation between a safety RS and safety risk. The second case analyses a capacity RS and explains the results by the system adaptation process through a multi-layer hierarchy.