Discovering and Evaluating Workflow Organizational Patterns from Events Log: An Agent based Approach

Abstract

This paper contributes to address an important issue in Workflow mining: organizational patterns mining issue. First, it reveals a critical and comparative study of three representative Workflow mining systems (InWolve, WorkflowMiner and ProM). The major drawback of these systems is their inability to deal with organizational patterns mining. This work considers organizational patterns as being social structures defining the activity distribution among actors involved in the Workflow, as well as the interaction protocols ruling the communications between them. To compensate the previous drawback, the paper proposes an agent based approach that includes an Events Log model integrating the interactions among actors using a performativebased enrichment of Events Log. This paper also gives the principles of organizational patterns mining and shows how evaluate the quality of discovered patterns and notably the organizational structures in terms of flexibility, efficiency and robustness. Finally, it describes DiscoopFlow that implements a crisis case study to validate the contributions. Discovering and Evaluating Workflow Organizational Patterns from Events Log: An Agent based Approach