Process mining
Also called: process discovery
Reconstructing how a process actually runs from the event logs its systems already produce, rather than from what people describe in workshops.
Process mining reconstructs the real behaviour of a process from the event logs its supporting systems already generate. Instead of asking people how they work — which produces the intended process — it reads timestamps and case identifiers and shows the paths actually taken, including the detours and rework nobody mentions in a workshop.
Its usual finding is the gap between the documented process and the operating one. That gap is diagnostic, not decorative: it identifies where the redesign should focus. Process mining describes; it does not execute. Its output is evidence for a decision, and the decision still has to be implemented by something that runs.
On this site: Governed AI →