Use multi-dimensional mining to identify inefficiencies and root causes to drive process excellence across multiple tables.

Before you begin

You must install and configure Process Mining plugins before configuring multi-dimensional mining on tables.

Role required: sn_process_optimization_analyst, sn_process_optimization_power_user, or sn_process_optimization_admin

About this task

Note: In setting up a project, you first must set up a main (parent) table configuration, define an activity definition, and set up a breakdown definition. If only one table is needed, you can generate a full model that you can periodically mine and schedule with the data. If multiple tables are required, you must create additional (child) tables, relate it to the main (parent) table, and define the activity definition.

Procedure

Create a project.
  1. Navigate to All > Process Mining > Analyst Workbench.
  2. Select Create New Project.
  3. Fill out the required fields and select Submit.
    The process model is created and defaulted as new.
Set up the multi-dimensional table relationship.
  1. Navigate to the Table Configuration tab and select New.
    A new table configuration new record page opens. relationship set up
  2. Select the Parent Table Configuration, Relationship, and Reference fields.
    The source type is set by default. Set the Table field and apply filters, if needed.new parent table configuration
    Note: Select Preview to see the number of records.
  3. Select Submit.
    You’re taken to a newly created Table Configuration page.
    Note: See Configure an activity definition and Configure a breakdown definition for more information.
Configure and create a child table.
  1. Navigate to the Table Configurations tab and select New.
    A New record Table Configuration page opens.
    1. Navigate to the Activity Definitions tab and select New.
    2. Enter a designated name in the Name field to configure the child table.
    3. Navigate to the Activity Definitions tab and select New.
    4. Fill in the required fields and select Submit.
    5. Link the current child table to the designated parent table.
      A reference field stores a reference to a field on another table. For example, the Problem field on the Incident table is a reference to the Problem table. Child records are related with a particular parent record. For example, requested items have service catalog tasks or incidents and SLA records.
    6. Select the search icon next to the Parent table configuration text box and select the parent multi-dimensional mining configuration.
    7. Select Reference or Child from the drop-down list, depending on the relation type.
      For child relation types, select Parent.
    8. Select Submit.
    A new child table configuration is created.

What to do next

You must finalize the configuration to see the full mining of your tables and their relation to one another.

Navigate to the Related links section and select Go to Project Definition. The parent and child tables appear in table configurations.

Multi-dimensional mining enables you to create several models and run your queries off records using that table. You can create relationships between different tables using multidimensional mining. You can mine multiple connected processes in one model in connected workflows.
Note: You can create additional parent and child tables by repeating the steps listed. For every table created, you must create an Activity Definition.

Your parent and child tables are configured and ready to be modeled. Select Generate Model (Sample). A pop-up displays the mining in progress. When the mining is successful, you can open Analyst Workbench to create and analyze the relationships between different tables. For the parent and child slider arc, a new double dotted line appears.

When you’re satisfied with the sample model, select Generate Model (Full).Analyst workbench view