Abstract: The use of latent variable competing risk models in reliability has at times been considered controversial due to the lack of identifiability without some added modelling assumptions. The introduction of a new framework of models, referred to as generalised competing risk models, provides the possibility that large classes of GCR models are identifiable. In this talk the GCR model framework is introduced, along with several diagrams to attempt to develop intuition about identifiability and the surrounding issues. An example class of GCR models will be introduced and a proof that this class is identifiable will be outlined.