This post is part of the Process Mining Use Case Series where we’ll focus on different techniques to identify process inefficiencies, non-conformant activities, and improvement opportunities. These posts will be broken into two sections – how to do the analysis and how to configure the Process Mining project.
Process mining is designed to help us get visibility into opportunities to improve our processes. But once we make those process improvements how do we know if they are having the expected impact and are those process changes having any adverse effects on other areas of the process? You can use KPIs to track the impact of the process change on things like closure times and satisfaction scores but getting visibility into adverse effects on other areas of the process will require Process Mining.
Process Mining can be used to do a side by side comparison and give us visibility into how a process is behaving both before and after a change so we know what’s working and what’s not.
How to do the Analysis
In this example we are going to use the condition filtering capability in the Analyst Workbench to create views of a process map for incidents that closed prior to a process change and after a process change. We’ll then be able to do a side by side comparison and analysis of the process before and after the process change.
This technique can be used on any process data, not just incident.
Open any Process Mining project to the Analyst Workbench view.
Select the Conditions option from the Advanced filters section that is at the bottom of the left hand panel (under the Breakdowns).
We can then use the Condition Builder to create the criteria to isolate the data prior to our process change. In my example we will be focusing on incidents that closed prior to October 31st.
Hit the Apply button.
If this is the first time you are mining for this criteria you will get the message that a task has been scheduled and you will need to pick it up from the scheduled tasks panel on the right hand side of the screen.
Once the mining is complete you’ll want to save this as a Saved Filter set.
Click on the “Your filter sets” option at the top of the panel on the left hand side of the screen.
Select “Save as new filter set”.
Enter a title for your filter set and click Save.
This view of the process will be now be available to use in single click.
Let’s bring the map back to its original state and create the view for after the process change.
Click on the Clear all button in the upper portion of the left hand panel (above the breakdowns).
Now to use the Condition Builder to isolate the process data for incidents closed after October 31st.
Click on the Conditions button in Advanced filters section in the lower left of the screen.
Use the Condition Builder to isolate incidents that have closed after October 31st.
Hit the Apply button.
Again if you have to wait for the mining to complete pick up your scheduled task on the right hand side of the screen.
Once the map has been isolated we will create saved filter set for after the process change.
Click on the “Your filter sets” button in the upper left and select “Save as new filter set”.
In the Save filter set dialogue enter “Process after the process change” as your title and click Save.
We now have our two views of the process saved and are set to be able to do a side by side comparison.
Click the Clear all button in the left hand panel to bring the Analyst Workbench to its original state.
Use the “Your filter sets” menu to select the “Prior to process change” filter set.
Now click on the “Compare” button in the upper right hand corner of the screen.
This will split the workbench into two halves. On the right hand side of the screen click on the “Clear all” button.
Now expand the “Show filters” option.
Use the “Your filter sets” option on the right to select your “Process after the process change” filter set.
We now have our side by side view of the process prior to the change and after the change and we can start to dig in to see how things are being affected.
If you want to get a view of the high level stats like volumes, route variants and durations you can use the “Comparison Statistics” option in the “Model Options” panel on the right hand side of the screen.
Click on the “Comparison Statistics” button.
You can see in my case that after the change the average duration of the process has decreased by 5 hours.
In this example we were using the condition builder and comparison feature to compare two different time periods. You can use the same technique to compare different regions, locations, departments, etc.
How to configure the project
There is no special Process Mining project configuration that needs to be applied to use this technique. Just be aware that if you are applying date filters in the analyst workbench you would want those dates to be within the date range or dates you configured at the project level.
If you have never created a project before use the video found here to help you get started..
If you are looking for more in-depth training you can use the Process Mining Academy library of content
You can find other Process Mining use cases here
