Alignment-Based Replayer(Default)

Alignment-based replay is the state-of-the-art technique in conformance checking. Alignments provide a robust and detailed view on the deviations. For each trace from the event log after replaying, there is one alignment corresponding to it.

  • Sync move
    The classification of the current event corresponds to the firing transitions in the Petri net. In this case, during replaying, both the trace and process model move to next comparison.
  • Move on log
    The classification of the current event doesn't have any corresponding firing transitions in the Petri net. During replaying, the trace moves forwards but the state of process model doesn't change.
  • Move on model
    The firing transitions in Petri net have no corresponding event in the trace. During replaying, the model moves forwards but the state of the event log doesn't change.

With node in Default Settting, the costs are unified for one move type and can't be set for each move.
More information seen in the : link

Options

Event classifer
The classifier chosen to classify the event log.
Strategy type
The strategy to replay the event log on Petri net. It includes
  • ILP Replayer.
    The repalyer calculates the optimal alignment based on ILP.
  • Non-ILP Replayer.
  • A*-ILP Based Manifest Replayer.
    This replayer is reserved for performance checking. It generates the Manifest to wrap the alignments.
Move on log cost
The unified cost for move on log in non-negative integer. In default, it is 1.
Move on model cost
The unified cost for move on model in non-negative integer. In default, it is 1.
Sync move cost
The unified cost for sync move in non-negative integer. In default, it is 0.

Input Ports

Icon
An event log as input
Icon
A Petri net as input

Output Ports

Icon
The alignment result after replaying.
To benefit the analysis later, the corresponding event log and Petri net are also implictly included in the result.

Views

This node has no views

Workflows

  • No workflows found

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.