Overview
The Time Period filter selects or removes cases based on date and time criteria. It provides flexible options for filtering cases using specific dates, relative time periods (such as "last month" or "this week"), or date ranges. The filter can operate at both the case level (filtering entire cases) and event level (filtering individual events within cases).
This filter supports multiple comparison methods including contained periods, intersecting periods, cases that started or completed within a period, and filtering based on specific activities or custom date attributes.
Common Uses
- Analyze process performance for the current quarter or fiscal year
- Compare cases from different time periods (e.g., this month vs. last month)
- Identify cases that started but haven't completed within a specific timeframe
- Filter out historical cases to focus on recent activity
- Analyze seasonal trends by selecting specific months or quarters
- Track active cases during a critical business period
- Exclude test or pilot cases that occurred before a specific go-live date
Settings
Start Date Mode: Defines the beginning boundary of the time period. You can choose from three options:
- Specific Date: Select an exact date and time. The filter will include cases on or after this date.
- Relative Period: Go back a specified number of days, weeks, months, or years from today. For example, "3 weeks" means the start date will be 3 weeks before today. This dynamically updates each day.
- Period Start: Select the start of a week, month, or year. You can specify how many periods to go back. For instance, "Start of Month" with "2" periods goes back to the first day of the month 2 months ago.
End Date Mode: Defines the ending boundary of the time period. Uses the same three options as Start Date Mode:
- Specific Date: Select an exact date and time. The filter will include cases on or before this date.
- Relative Period: Go forward or back a specified number of days, weeks, months, or years from today.
- Period End: Select the end of a week, month, or year. For instance, "End of Month" with "1" period means the last day of last month.
Time Period: Specifies the number of periods to go back (or forward into the future) when using relative or period-based date modes. For example, entering "3" with "Months" means 3 months back from today.
Filter Type: Determines how cases are selected based on their relationship to the time period:
- Active: Selects cases that had any activity during the selected period. A case is active if it overlaps with the time period in any way.
- Activity: Selects cases where a specific activity occurred during the period. Requires you to specify which activity to check.
- Attribute: Selects cases where a custom date attribute value falls within the period. Useful for filtering based on due dates, approval dates, or other date fields.
- Completed: Selects cases that finished during the selected period (case end time within the range).
- Contained: Selects cases that both started and finished entirely within the selected period. Both case start and end times must be within the range.
- Events: Filters individual events within the time period and returns cases containing those events. This is event-level filtering rather than case-level.
- Intersecting: Selects cases that overlap with the time period in any way. Similar to Active but may use different date comparison logic.
- Started: Selects cases that began during the selected period (case start time within the range).
Activity (conditionally required): When Filter Type is set to "Activity", specify which activity to check for occurrence during the time period. The filter will select cases where this activity happened within the date range.
Attribute (conditionally required): When Filter Type is set to "Attribute", specify which date/time attribute to filter on. This must be a valid date attribute in your process data.
Remove Selected Cases: When checked, inverts the filter logic to exclude matching cases instead of including them. Use this to remove cases from your analysis rather than selecting them.
Examples
Example 1: Completed Cases in 2024
Scenario: You want to analyze all cases that were completed during the 2024 calendar year to understand annual performance.
Settings:
- Start Date Mode: Specific Date = January 1, 2024 00:00:00
- End Date Mode: Specific Date = December 31, 2024 23:59:59
- Filter Type: Completed
- Remove Selected Cases: Unchecked
Result: The filter selects all cases where the case end time falls between January 1, 2024 and December 31, 2024. Cases that started in 2023 but finished in 2024 are included. Cases that started in 2024 but haven't finished yet are excluded.
Insights: This helps you measure completed work for annual reporting, identify completion trends throughout the year, and calculate accurate throughput metrics for the full calendar year.
Example 2: Cases Active Last Month
Scenario: You need to review all cases that had any activity during the previous month for monthly operational reporting.
Settings:
- Start Date Mode: Period Start = Start of Month, Time Period = 1 (past)
- End Date Mode: Period End = End of Month, Time Period = 1 (past)
- Filter Type: Active
- Remove Selected Cases: Unchecked
Result: The filter dynamically selects cases that were active during last month. If today is March 15th, it selects cases active during February 1st through February 28th (or 29th). Tomorrow it will still show last month's data.
Insights: This provides a rolling monthly view of active work. Use it to track monthly workload, identify cases that span multiple months, and monitor ongoing work volumes. The dynamic nature means the same filter configuration works month after month.
Example 3: Recent Orders (Last 30 Days)
Scenario: You want to focus your analysis on recent activity by filtering cases where orders were created in the last 30 days.
Settings:
- Start Date Mode: Relative Period = Days, Time Period = 30 (past)
- End Date Mode: (leave empty/null to include all dates after the start)
- Filter Type: Activity
- Activity: Create Order
- Remove Selected Cases: Unchecked
Result: The filter selects cases where the "Create Order" activity occurred within the last 30 days. If today is June 15th, it includes cases where "Create Order" happened between May 16th and June 15th. Tomorrow the window shifts forward by one day.
Insights: This creates a rolling 30-day window focused on recent order creation. Use it to monitor recent order processing, identify current bottlenecks, and track real-time process performance without including historical data that might skew your analysis.
Example 4: Q1 2024 Started Cases
Scenario: You want to analyze cases that started during the first quarter of 2024 to evaluate the beginning of year performance.
Settings:
- Start Date Mode: Specific Date = January 1, 2024 00:00:00
- End Date Mode: Specific Date = March 31, 2024 23:59:59
- Filter Type: Started
- Remove Selected Cases: Unchecked
Result: The filter selects all cases that began between January 1 and March 31, 2024. Cases may still be in progress or may have completed later in the year.
Insights: This helps analyze how work initiated during Q1 progressed, compare quarterly intake volumes, and assess whether Q1 cases met expected timelines.
Example 5: Exclude Old Test Cases
Scenario: Your process went live on July 1, 2024, and you want to exclude all test cases that occurred before the go-live date.
Settings:
- Start Date Mode: (leave empty/null)
- End Date Mode: Specific Date = June 30, 2024 23:59:59
- Filter Type: Completed
- Remove Selected Cases: Checked (important!)
Result: The filter removes all cases that completed on or before June 30, 2024. Only cases completed after the go-live date remain in your dataset.
Insights: This cleanses your dataset by removing test and pre-production cases, ensuring your analysis reflects actual production process performance without contamination from test data.
Example 6: Invoices Approved This Week
Scenario: You want to monitor invoices that were approved during the current week to track weekly approval volumes.
Settings:
- Start Date Mode: Period Start = Start of Week, Time Period = 0 (current week)
- End Date Mode: Period End = End of Week, Time Period = 0 (current week)
- Filter Type: Activity
- Activity: Approve Invoice
- Remove Selected Cases: Unchecked
Result: The filter selects cases where "Approve Invoice" occurred during the current week (Monday 00:00:00 to Sunday 23:59:59 of the current week).
Insights: This provides a rolling weekly view of approval activity. Use it to track weekly productivity, identify approval bottlenecks during the week, and monitor whether approvals are keeping pace with incoming invoices.
Output
Cases Included/Excluded: Based on the Filter Type and date range, the filter either includes or excludes complete cases from your dataset.
Event-Level Filtering: When Filter Type is set to "Events", the filter operates on individual events rather than complete cases. Cases are reconstructed to include only events within the specified time period.
Dynamic Updates: Filters using relative time periods (like "last 30 days" or "this month") automatically update each day. The same filter configuration produces different results tomorrow because the reference point (today) has shifted.
Case Timeline Impact: The filter affects case timelines differently based on Filter Type:
- "Contained" requires both start and end within the period
- "Started" only checks case start time
- "Completed" only checks case end time
- "Active" checks for any overlap between case duration and the time period
- "Intersecting" checks for any overlap with potentially different comparison logic than Active
Attribute Filtering: When using Filter Type "Attribute", the filter checks custom date fields rather than case start/end times. This allows filtering based on business-specific dates like due dates, approval dates, or delivery dates.
Inverted Logic: When "Remove Selected Cases" is checked, all matching cases are excluded rather than included. This is useful for removing outliers, test data, or cases from specific problematic periods.
This documentation is part of the mindzie Studio process mining platform.