A complex system, be it a manufacturing system or otherwise, is prone to abnormal functioning, as its units or elements experience unpredictable functional variation. Determining the likely root causes of its undesirable functional events is required to carry out cause analysis of the system from functional point of view. Although the manufacturers provide some information in the manual in this regard, yet these are not conclusive and fall much short in guiding the users in functional cause analysis. Structure plays an important role in this objective. In this article, a procedure for functional cause analysis of a complex manufacturing system through structure is presented using digraph models. ‘Function event digraph’ is defined for a complex manufacturing system at its various hierarchical levels by considering its input and output functions and their interrelations. A top or undesired functional event for the system is identified from its digraph model, and its root causes are obtained by developing ‘functional cause analysis tree’. The suggested approach helps designers and practicing engineers in functional cause analysis of manufacturing systems and thus leading to reliability enhancement and sustenance. An example of a complex CNC (computer numerical control) grinding machine is illustrated to demonstrate the methodology.