Process discovery
Process discovery is the process-mining technique of constructing a process model (typically BPMN or a Petri net) from an event log without prior knowledge of the intended process. Discovery algorithms — Alpha Miner, Heuristic Miner, Inductive Miner — produce the model that best explains the observed event sequences.
The output of process discovery is often surprising: the discovered process diverges from the documented process in ways nobody noticed. Common discoveries: the 'happy path' the documentation describes happens in only 30% of cases; a 'rare exception' handler runs for 25% of cases; loops the team didn't realise existed account for hours of cycle time. The pragmatic use: discover, identify the top 2-3 variants by frequency, ask 'is this what we actually want?', and either change the process or update the documentation. Discovery is the first step; conformance checking, variant analysis, and enhancement follow.
Discussed in our use-cases
ICP-targeted pages where process discovery is part of the framing.
Related terms
- Conformance checking
Conformance checking is the formal process-mining technique of comparing an event log to a reference process model and producing fitness, precision, generalisation, and simplicity scores.
- Variant analysis
Variant analysis groups cases by the unique sequence of activities they followed and ranks variants by frequency and cost.
- Event log
An event log is the chronological record of activities executed in a process, structured so each row represents one event (case ID, activity name, timestamp, optional resource).