In Process Mining, Model Enhancement describes the analysis of a data-driven process model for optimization potential. The data-driven process model is based on a log file of the process. Based on this information, improvements or changes are made. For example, bottlenecks or unplanned process sequences identified in this way can be eliminated. The objective of Model Enhancement is the optimization of the process model and thus of the underlying process.
Creation of a new process model.
Why is Model Enhancement so important?
The results of the Model Enhancement reflect the quality of the analysis and are the reference for future analysis. In Model Enhancement, the results are used by the Process Discovery and Conformance Checking analysis techniques. The correct implementation of identified optimization potentials is essential: If a company makes unfavorable process changes, this can have costly consequences. For example, this can result in higher expenses, rule violations or quality defects. And until the responsible process change has been identified and eliminated, it usually takes some time.
What exactly does Model Enhancement do?
Let’s start at the beginning. Suppose we perform a Process Mining analysis with the log file of any business process. Using the Process Discovery and Conformance Checking methods, we identify various process weaknesses: These include bottlenecks, process loops and unwanted process deviations. This is all valuable information. Now we know where to find which types of optimization potential.
But what do we do with this knowledge? We specifically adapt our target model, which serves as the standard and guideline for process implementation. This means for instance that we change the process in such a way that the risk of bottlenecks is reduced or certain process sequences are no longer possible. The success or usefulness of the process changes can of course only be determined after the new process has been implemented for some time. Therefore, Process Mining is an excellent way to continuously improve your process (see CIP). Because the changes can be more or less successful.
In practice, processes are usually subject to step-by-step, marginal changes.
This is how it works:
1. Analysis of the process data
2. Identification of optimization potentials (Process Discovery, Conformance Checking)
3. Adaptation of the target process model (Model Enhancement)
4. Implementation of the target process model
5. Checking of the new process implementation on the basis of the analysis of the process data (continuous improvement process)