Overview
The Events with Attribute filter selects or removes individual events based on attribute values. This powerful filter operates at the event level, allowing you to filter by both event-level attributes (like activity name, resource, or timestamp) and case-level attributes (like customer type or region). It supports filtering by text, numbers, dates, durations, and boolean values using various comparison methods.
Unlike case-level filters that include or exclude entire cases, this filter evaluates each event individually and only affects the specific events that match your criteria - unless you filter by a case-level attribute, in which case all events from matching cases are affected.
Common Uses
- Select all "Create Invoice" activity events to analyze invoice creation patterns
- Filter events performed by specific resources to review workload distribution
- Exclude events that occurred outside business hours using timestamp filtering
- Select high-value transaction events where item cost exceeds a threshold
- Find events with specific attribute values to investigate process variations
- Remove test data by filtering out events with test user identifiers
Settings
Attribute: Select the attribute (column) you want to filter on. This can be any event-level attribute (like Activity Name, Resource, or Event Time) or case-level attribute (like Customer Type or Region). The filter automatically detects whether the attribute is at the event or case level.
Compare Type: Choose how to compare the attribute values. Available options depend on the data type of your selected attribute:
- Text attributes: Equal, Begins With, Ends With, Contains, Is One Of
- Numeric attributes: Equal, Greater Than, Greater Than or Equal, Less Than, Less Than or Equal, Is One Of
- Date/time attributes: Equal, Greater Than, Greater Than or Equal, Less Than, Less Than or Equal, Day Equal, Day Greater Than, Day Greater Than or Equal, Day Less Than, Day Less Than or Equal
- Duration attributes: Same as numeric, with additional time unit selection (Days, Hours, Minutes, Seconds, Milliseconds)
- Boolean attributes: Equal (True or False)
Compare Value: Specify the value to compare against. For "Is One Of" comparisons, you can select multiple values from a list. The value type must match your selected attribute's data type.
Time Units: (For duration attributes only) Specify the time unit for comparison - Days, Hours, Minutes, Seconds, or Milliseconds.
Remove Selected Events: When checked, the filter excludes events that match your criteria instead of including them. This inverts the filter logic without changing your comparison settings.
Examples
Example 1: Filter by Activity Name
Scenario: You want to analyze only the "Create Order" events in your order-to-cash process to understand order creation patterns.
Settings:
- Attribute: Activity Name
- Compare Type: Equal
- Compare Value: Create Order
- Remove Selected Events: Unchecked
Result: The filter selects all events where the activity is "Create Order". All other activity events are excluded from the analysis.
Insights: This allows you to focus your analysis on order creation, examine the resources involved, check timing patterns, and analyze any attributes specific to order creation events.
Example 2: Select Events by Multiple Resources
Scenario: You want to review the work performed by your senior team members: John Doe, Jane Smith, and Bob Wilson.
Settings:
- Attribute: Resource
- Compare Type: Is One Of
- Compare Value: John Doe, Jane Smith, Bob Wilson
- Remove Selected Events: Unchecked
Result: The filter selects all events performed by any of the three specified resources. Events performed by other resources are excluded.
Insights: This helps you analyze the activities, throughput, and performance patterns of your senior team members compared to the rest of the team.
Example 3: Filter High-Value Transaction Events
Scenario: You want to focus on events involving transactions over $1,000 to analyze high-value order handling.
Settings:
- Attribute: Item Cost
- Compare Type: Greater Than
- Compare Value: 1000
- Remove Selected Events: Unchecked
Result: Only events where the item cost exceeds $1,000 are selected for analysis.
Insights: This reveals how high-value transactions are processed differently from regular transactions, including which resources handle them, how long they take, and what additional steps might be involved.
Example 4: Exclude Events Outside Business Hours
Scenario: You want to analyze only events that occurred during business hours (before 6 PM) to exclude after-hours processing.
Settings:
- Attribute: Event Time
- Compare Type: Less Than or Equal
- Compare Value: 18:00:00 (6 PM)
- Remove Selected Events: Unchecked
Result: Only events that occurred at or before 6 PM are included in the analysis.
Insights: This helps you understand the normal business-hours process flow without the noise of after-hours exceptions or batch processing.
Example 5: Filter Events by Case-Level Attribute
Scenario: You want to analyze events only from cases belonging to Premium customers.
Settings:
- Attribute: Customer Type
- Compare Type: Equal
- Compare Value: Premium
- Remove Selected Events: Unchecked
Result: All events from cases where Customer Type equals "Premium" are selected. This includes all activities in those cases, even though you're filtering by a case-level attribute.
Insights: When filtering by case-level attributes, the filter affects all events in matching cases, allowing you to see the complete process flow for specific case types.
Example 6: Select Events by Date Range
Scenario: You want to analyze only events that occurred in January 2024 to compare monthly performance.
Settings:
- Attribute: Event Time
- Compare Type: Day Greater Than or Equal
- Compare Value: 2024-01-01
- Additional filter needed: Day Less Than, value 2024-02-01
Result: Only events that occurred on dates from January 1-31, 2024 are selected (ignoring the time portion).
Insights: Using the Day comparison methods allows you to filter by calendar dates without worrying about exact timestamps, making date-range analysis simpler.
Output
After applying the Events with Attribute filter:
- Events that match your criteria are included in the filtered event log (or excluded if "Remove Selected Events" is checked)
- Cases that no longer have any events are completely removed from the analysis
- Cases that still have some events remain, even if they lost some events due to filtering
- Your process map, statistics, and other analyses update to reflect only the filtered events
- The case count shown in mindzieStudio reflects the number of cases that still have at least one event after filtering
Important: Unlike case-level filters, this filter can result in partial cases - cases that have some events removed but still appear in the analysis with their remaining events. This is particularly useful when you want to analyze specific activities or events while maintaining the context of the cases they belong to.
This documentation is part of the mindzie Studio process mining platform.