Minit 3.1: Subprocess Analysis Use Case
Is your process too big and too complicated? Minit 3.1 introduces new event-level filtering capabilities with the best-in-breed performance and user experience. You can easily filter out the noise and focus only on that part of the process that is currently important to you, without manually preprocessing the data before the import to do so.
Every Process Analysis needs to start with clear questions you want to answer. And many more will come up as soon as you start digging into to structure of your process. But what if you have never seen your process data in a structured way and you don't have any process models at all? Where do you begin?
With the right tool, you can get first process insights quickly and easily. At the beginning of your project, Minit Process Discovery capabilities will help you reconstruct the process from the traces left within the IT system where the process takes place. This discovery technique can be used to both map and to document the process and visualize the as-is process model.
If the processes at your company are not precisely prescribed, right after the first data import you will be dazzled and at the same time confused when your process mining software visualizes the process as the „spaghetti” diagram. As you can see, when we include all event types in this real-life process of Loan application you get lost in an incredibly complex process model. Even after filtering out some less frequent activities and connections it's still not easily readable. With new Minit Subprocess analysis feature, you can quickly focus on the critical parts of the process.
In our example, we filtered out all Offer related events and Workitem associated events. The remaining „A“ events give us an excellent big picture overview of the application process itself. And that's about it!
We can see that 35% of all applications skip the A Submitted event. The bank starts these applications. They are probably coming in via a different channel, such as an advisor in a branch, and not the customer-facing online systém.
23. 05. 2018