process-mining
latest
false
UiPath logo, featuring letters U and I in white

Process Mining

Automation CloudAutomation Cloud Public SectorAutomation SuiteStandalone
Last updated Dec 3, 2024

Starting a Task Mining project from Process Mining

Introduction

Different UiPath® products work together to discover, investigate and build automation and communicate process improvements. Task Mining is an AI driven discovery product. It collects employee desktop data comprising screenshots and logs data upon each user action (mouse click, keystroke). Then it leverages AI to analyze the data.

Process Mining and Task Mining

With Process Mining, you extract the data from your business applications to get a detailed understanding of overall business processes. From Process Mining, you can zoom in on certain activities with Task Mining to gain deeper understanding of the disparate applications used to perform the activity (in addition to the system of record as observed in Process Mining) on employee machines.

Starting a Task Mining project from Process Mining

When you identify a bottleneck in your process while analyzing or want to get more granular visibility, you can start a Task Mining project from any process graph in Process Mining. This enables you to get an end-to-end view of your process including system and desktop level interactions along with the different applications used to perform the activity. This can help you to better understand task level drill down of activities during the process execution.

Note:

You can also start a Task Mining project when performing a Root Cause Analysis or Simulating Automation Potential.

For more information on Task Mining, see the Task Mining documentation.

Prerequisites

The Start new Task Mining project option in Process Mining will be enabled when the following prerequisits are met.

  • The Task Mining service must be enabled on the same tenant as Process Mining.
  • You need access rights in Task Mining to create Task Mining projects.
  • You must have the Administrator or Project Creator role in Task Mining.

Starting a Task Mining project

Follow these steps to start a Task Mining project.

  1. In the process graph, select the activity for which you want to see task level information.

  2. Select Task Mining ->Submit to Task Mining project from the context menu.

    The Submit to Task Mining wizard is displayed at the right.


  3. Enter a name for the Task Mining project in the Project name field.

  4. If desired, change the Description. By default, the name of the selected activity and the link to the Task Mining project that will be created are displayed in the Description field.

  5. Select Next to go to the Invite users step.

  6. Add recording users by typing the username, group name, or email and select the desired result from the dropdown list.

    Note:

    Users will receive an email invitation to the project and a link to download the Task Mining recorder.

  7. Select Create to create the Task Mining project and to start the recording. You can select the Task Mining project link to view the Task Mining project and to analyze the results.


Viewing Task Mining projects

You can view the Task Mining projects that were created for the activity directly from the process graph menu and open the project in Task Mining.

Note:

If a Task Mining project has been submitted from an activity, this is indicated by the Task Mining icon displayed on that particular activity.

Follow these steps to view the Task Mining projects started from your process.

  1. In the process graph, select Task Mining -> View Projects from the menu.

    The Task Mining projects panel is displayed, showing the Task Mining project cards for the projects created from the process app.

  2. Locate the project card for the Task Mining project you want to view and select View data.

The Task Mining trace diagram for the project is shown alongside key KPIs, which provide trace details such as application usage, number of steps, and comparisons between waiting and cycle time.

Note:

The trace diagram displays the trace that is marked as the Ideal trace in the Task Mining project. The Ideal trace is the Task Mining trace that is representative of the applicable Process Mining activity scenario.

Refer to Trace details in the Task Mining guide for more information on the ideal trace.

Task Mining trace

The Task Mining trace displays the sequence of steps taken to complete the particular activity. Each step in the trace includes the title or description of the step, its duration, waiting time, and the application that was used. This provides a detailed view of the activity, that can be used to analyze and optimize the performance of the activity.


Trace details

The Trace details panel provides more detailed information about the activity trace.

Below is a description of the KPIs.

Metric

Description

Number of steps

The total number of steps in the trace.

Event throughput time (expected vs real)

The total execution time of the activity.
Note:

The actual amount of time it took to execute the activity, including waiting time, is displayed in red.

Waiting time

The total waiting time for the activity.

Applications

The applications/domains recorded in the trace.
Note:

For each application/domain the percentage of total event throughput time is displayed.

View in Task Mining

In Task Mining, you can combine similar step sequences recorded from different instances of an activity and merge them into a merged trace.

By merging these similar traces, you can get a clearer, high-level view of the process, which should make it easier to identify common patterns, bottlenecks, and opportunities for automation or process improvement.

Select the View in Task Mining option to open the project in Task Mining and select the traces you want to merge.

Refer to Merge traces in the Task Mining guide for more information on merging traces.

Was this page helpful?

Get The Help You Need
Learning RPA - Automation Courses
UiPath Community Forum
Uipath Logo White
Trust and Security
© 2005-2024 UiPath. All rights reserved.