Cascading delays that propagate from a primary source along a railway network is an immediate concern for British railway systems. Complex non-linear interactions between various spatio-temporal variables govern the propagation of these delays which can avalanche throughout railways networks causing further severe disruptions. This paper introduces several machine learning techniques alongside data mining processes to create a framework that forecasts key performance indicators (KPIs). Not only do these frameworks provide great accuracy, they also allow for insight into understanding the mechanism of delay propagation with the railway network. This paper goes on to perform up to ten future steps for KPI predictions through state-of-the-art ML models for trains at a certain checkpoint. Further discussion on the improvements, applicability and scalability of this framework are also provided in this paper.