Overview
The Cases with Attribute filter selects or removes cases based on attribute values at either the case level or event level. This versatile filter supports comprehensive comparison operations including exact matching, text pattern matching, numerical comparisons, date operations, and multi-value selections.
The filter automatically detects whether the specified attribute exists in the case data or event data and applies the appropriate filtering logic. When filtering on event attributes, entire cases are included or excluded based on whether they contain events matching the criteria - the filter does not remove individual events.
Common Uses
Include cases where:
- All cases from a specific region or department
- Cases with total order value exceeding a threshold
- Cases containing a specific activity or resource
- Cases started or completed within a date range
- Cases where a boolean flag is true (e.g., "Expedited" or "Cancelled")
- Cases with customer type matching one of several categories
Exclude cases where:
- Cases from test accounts or inactive vendors
- Cases with amounts below minimum processing thresholds
- Cases that don't contain required activities
- Cases outside your analysis timeframe
- Cases marked as cancelled or invalid
- Cases from specific organizational units being reorganized
Settings
Attribute: Select the attribute you want to filter on from the dropdown menu. This can be any case-level attribute (like Region, Customer, Total Amount) or event-level attribute (like Activity Name, Resource, Event Status). The filter automatically determines whether the attribute exists at the case or event level.
Comparison Method: Choose how to compare attribute values against your criteria. Available comparison methods depend on the attribute's data type:
- Text attributes: Equal, Not Equal, Begins With, Ends With, Contains, Is One Of
- Numeric attributes: Equal, Not Equal, Greater Than, Greater Than or Equal, Less Than, Less Than or Equal, Between
- Date attributes: Day Equal, Day Greater Than, Day Greater Than or Equal, Day Less Than, Day Less Than or Equal, Between
- Boolean attributes: Equal (True/False)
Compare Value: Enter the value to compare against when using single-value comparisons (Equal, Not Equal, Greater Than, etc.). For text comparisons, matching is case-insensitive.
Compare Values (Is One Of): When using "Is One Of" comparison, select multiple values from the list or enter multiple values. Cases matching any of the specified values will be included.
Range Values (Between): For Between comparisons, specify both the lower and upper bounds. The comparison is inclusive - cases with values equal to either bound are included.
Activity Filter (Event Attributes Only): When filtering on event-level attributes, you can optionally specify an activity name to limit the evaluation to events from that specific activity. If left blank, all events are considered regardless of their activity.
Remove Selected Cases: Check this box to invert the filter logic - instead of including cases that match the criteria, the filter will exclude them. This is useful for removing unwanted cases from your analysis.
Examples
Example 1: Filter by Region (Text Exact Match)
Scenario: You want to analyze only cases from the European region to compare performance against other regions.
Settings:
- Attribute: Region
- Comparison Method: Equal
- Compare Value: "Europe"
- Remove Selected Cases: Unchecked
Result:
- Cases Included: All cases where Region = "Europe"
- Cases Excluded: All cases from other regions (North America, Asia, etc.)
Insights: This creates a focused dataset for regional analysis, allowing you to identify region-specific patterns and compare metrics against other regions.
Example 2: Filter High-Value Orders (Numeric Comparison)
Scenario: You want to focus on high-value purchase orders exceeding $50,000 to analyze approval patterns and processing times.
Settings:
- Attribute: Total Order Amount
- Comparison Method: Greater Than
- Compare Value: 50000
- Remove Selected Cases: Unchecked
Result:
- Cases Included: All cases where Total Order Amount > $50,000
- Cases Excluded: All cases with amounts of $50,000 or less
Insights: Isolating high-value transactions helps identify whether approval bottlenecks or compliance issues are specific to large orders, enabling targeted process improvements.
Example 3: Filter by Multiple Vendor Categories (Multi-Value Selection)
Scenario: You need to analyze cases from preferred vendors in categories A, B, and C while excluding other vendor categories.
Settings:
- Attribute: Vendor Category
- Comparison Method: Is One Of
- Compare Values: ["Category A", "Category B", "Category C"]
- Remove Selected Cases: Unchecked
Result:
- Cases Included: All cases where Vendor Category is A, B, or C
- Cases Excluded: All cases from other vendor categories
Insights: This allows focused analysis on preferred vendor performance while maintaining a sufficient case volume for meaningful insights across multiple related categories.
Example 4: Find Cases with Specific Activity (Event Attribute)
Scenario: You want to find all cases that went through manual approval to understand how often this exception path occurs.
Settings:
- Attribute: Activity Name
- Comparison Method: Equal
- Compare Value: "Manual Approval"
- Remove Selected Cases: Unchecked
Result:
- Cases Included: All cases containing at least one "Manual Approval" event
- Cases Excluded: All cases that never had manual approval
Insights: Identifying cases requiring manual approval helps quantify automation rates and understand which types of cases require human intervention.
Example 5: Filter Recent Cases (Date Range)
Scenario: You want to analyze only cases that started in the last quarter for a current performance assessment.
Settings:
- Attribute: Case Start Date
- Comparison Method: Between
- Lower Value: 2024-07-01
- Upper Value: 2024-09-30
- Remove Selected Cases: Unchecked
Result:
- Cases Included: All cases with start dates from July 1 through September 30, 2024
- Cases Excluded: All cases started before July 1 or after September 30, 2024
Insights: Time-based filtering ensures your analysis reflects current process performance rather than historical patterns that may no longer be relevant.
Example 6: Exclude Test Accounts (Inverse Filter)
Scenario: Your dataset includes test cases that should not be part of operational analysis. You want to remove all cases where the "Test Account" flag is true.
Settings:
- Attribute: Is Test Account
- Comparison Method: Equal
- Compare Value: True
- Remove Selected Cases: Checked
Result:
- Cases Included: All cases where Is Test Account = False or Null
- Cases Excluded: All cases where Is Test Account = True
Insights: Removing test data ensures your metrics and analysis reflect actual operational performance and aren't skewed by testing activities.
Output
The filter modifies the case selection in your current analysis view. The case count indicator at the top of the screen updates to show how many cases remain after filtering. All subsequent calculators, visualizations, and analysis tools will operate only on the filtered case set.
The filter operates at the case level - even when filtering on event attributes, entire cases are included or excluded based on whether they contain matching events. Individual events are not removed from cases.
When multiple filters are applied, they work together using AND logic - a case must pass all filter criteria to be included in the analysis.
This documentation is part of the mindzie Studio process mining platform.