Recently, a new generation of adaptive Process-Aware Information Systems (PAIS) has emerged, which enables dynamic process changes (e.g., to insert, delete, and move activities). This, in turn, has led to a large number of process variants derived from the same reference model, but differing in structure due to the applied changes. Generally, such process variants are expensive to configure and difficult to maintain. In this talk, we will introduce a heuristic search algorithm to discover a reference process model by mining these variants. The discovered model is expected to have smaller average distance to the variants so that less configuration efforts will become necessary in future, if this model is adopted as the reference model. Our algorithm provides additional advantages; e.g., it can also take the original reference process model into consideration. Such function allows us to control the similarity between the discovered model and the original reference model by setting different search distances. Through this approach we can limit the efforts for re-designing the reference model and PAIS respectively. Additionally, we can avoid spagitti-like process models which are too difficult to understand.
Heuristic Process Variants Mining
Ulm University Ulm University![Logo](https://www.uni-ulm.de/_assets/a92153751098915699a1afa17e77f864/Images/logo-uni-ulm.jpg)
Gastvortrag, Chen Li (University of Twente, Niederlande), Ort: O27/545, Zeit: 11:00 Uhr, Datum: 18. Dezember 2008