Overview
The Shift Activity Time enrichment is a powerful time adjustment tool that allows you to modify the timestamps of specific activities within your process event log. This enrichment shifts activity timestamps forward or backward by a specified number of hours, making it invaluable for time zone corrections, data alignment across different systems, or simulation scenarios.
This enrichment operates at the activity level, meaning you can target specific activities for time adjustment while leaving others unchanged. It includes intelligent handling for date-only values (timestamps at midnight), which can be preserved to maintain data integrity when working with systems that store dates without time components. The enrichment directly modifies the event timestamps in your dataset, enabling proper chronological analysis and accurate duration calculations after time zone adjustments or system clock corrections.
Common Uses
- Time Zone Corrections: Adjust timestamps when activities are recorded in different time zones, ensuring all events align to a single reference time zone for accurate process analysis
- System Clock Misalignment: Correct timestamp discrepancies when different systems have clock synchronization issues or record events with systematic time offsets
- Data Integration: Align timestamps when merging event logs from multiple sources that use different time references or recording conventions
- Daylight Saving Time Adjustments: Compensate for daylight saving time changes that may affect year-over-year comparisons or seasonal analysis
- Simulation and Testing: Create what-if scenarios by shifting activities to different time periods to test process behavior under various timing conditions
- Legacy Data Correction: Fix historical data where timestamps were recorded incorrectly due to system configuration errors or data migration issues
- Cross-Regional Process Analysis: Normalize timestamps for global processes where activities occur across multiple time zones but need unified analysis
Settings
Activity Names: Select the specific activity whose timestamps you want to adjust. The dropdown presents all activities found in your current dataset. Only events with the selected activity name will have their timestamps modified, while all other activities remain unchanged. This targeted approach allows precise control over which parts of your process are time-shifted.
Shift Hours: Enter the number of hours to shift the selected activity's timestamps. Positive values move timestamps forward (into the future), while negative values move them backward (into the past). For example, entering 5 will add 5 hours to each occurrence of the selected activity, while -3 will subtract 3 hours. This accepts decimal values for sub-hour adjustments (e.g., 1.5 for 90 minutes).
Don't Shift Midnight: When enabled (default), timestamps that occur exactly at midnight (00:00:00) are not adjusted. This setting preserves date-only values that many systems store as midnight timestamps when time information is not relevant. Disable this option if you need to shift all timestamps regardless of their time component, such as when performing a uniform time zone conversion where midnight timestamps are genuine time values.
Examples
Example 1: Correcting Time Zone Misalignment in Global Supply Chain
Scenario: A multinational company's order processing system records "Order Placed" activities in UTC, but the warehouse management system records "Shipment Dispatched" in local Eastern Time (UTC-5). This 5-hour difference makes it appear that some shipments occur before orders are placed.
Settings:
- Activity Names: Shipment Dispatched
- Shift Hours: 5
- Don't Shift Midnight: Enabled
Output: The enrichment adjusts all "Shipment Dispatched" timestamps by adding 5 hours, converting them from Eastern Time to UTC. For example:
- Original "Shipment Dispatched": 2024-03-15 09:30:00 (ET)
- Adjusted "Shipment Dispatched": 2024-03-15 14:30:00 (UTC)
- "Order Placed" remains: 2024-03-15 13:45:00 (UTC)
The process flow now shows the correct sequence with orders preceding shipments.
Insights: After correction, the true lead time from order to shipment is revealed as 45 minutes instead of appearing as a negative duration. This enables accurate KPI calculation and identifies genuine process bottlenecks.
Example 2: Adjusting for Daylight Saving Time in Healthcare Scheduling
Scenario: A hospital's patient scheduling system didn't properly account for the spring daylight saving time change. All "Appointment Scheduled" activities for a two-week period in March need to be shifted forward by one hour to reflect the actual appointment times patients received.
Settings:
- Activity Names: Appointment Scheduled
- Shift Hours: 1
- Don't Shift Midnight: Enabled
Output: The enrichment shifts all "Appointment Scheduled" events forward by one hour:
- Original timestamp: 2024-03-12 10:00:00
- Adjusted timestamp: 2024-03-12 11:00:00
- Other activities like "Patient Arrived" and "Consultation Started" remain unchanged
This correction ensures appointment scheduling metrics accurately reflect the intended appointment times.
Insights: The adjustment reveals that the apparent increase in patient no-shows during this period was actually due to the time recording error, not actual patient behavior changes.
Example 3: Integrating Legacy System Data in Manufacturing
Scenario: A manufacturing plant is integrating historical data from an old production line system that recorded all timestamps 8 hours behind due to incorrect time zone configuration. All "Quality Check Completed" activities from this system need correction before process mining analysis.
Settings:
- Activity Names: Quality Check Completed
- Shift Hours: 8
- Don't Shift Midnight: Disabled
Output: All "Quality Check Completed" timestamps are shifted forward by 8 hours:
- Original: 2024-01-15 00:00:00 (midnight timestamp)
- Adjusted: 2024-01-15 08:00:00 (correctly shifted even though it was midnight)
- Manufacturing events from the modern system remain unchanged
With "Don't Shift Midnight" disabled, even date-only values stored as midnight are properly adjusted.
Insights: The corrected data shows quality checks actually occurred during the day shift as expected, not during unmanned overnight periods, enabling accurate shift performance analysis.
Example 4: Simulating Process Changes in Financial Services
Scenario: A bank wants to simulate the impact of moving their overnight batch processing 3 hours earlier to reduce morning system load. They need to shift all "Batch Processing Started" activities backward to analyze potential conflicts with end-of-day operations.
Settings:
- Activity Names: Batch Processing Started
- Shift Hours: -3
- Don't Shift Midnight: Disabled
Output: The enrichment moves all batch processing starts 3 hours earlier:
- Original start time: 2024-02-20 02:00:00
- Simulated start time: 2024-02-19 23:00:00 (shifts to previous day)
- All downstream activities remain at original times for impact analysis
Insights: The simulation reveals that starting batch processing 3 hours earlier would create conflicts with end-of-day reconciliation processes that run until 11:30 PM, requiring process redesign before implementation.
Example 5: Correcting Cross-Regional Data in Procurement
Scenario: A procurement system records "PO Approved" activities from the European office in CET (Central European Time) while the US office records them in PST (Pacific Standard Time). To analyze the global approval process, all US approvals need to be converted to CET by adding 9 hours.
Settings:
- Activity Names: PO Approved
- Shift Hours: 9
- Don't Shift Midnight: Enabled
Output: US-based PO approvals are shifted from PST to CET:
- Original US approval: 2024-04-10 14:00:00 PST
- Adjusted to CET: 2024-04-10 23:00:00 CET
- European approvals remain unchanged
- Purchase orders with date-only approval timestamps (stored as midnight) are preserved
Insights: The unified timeline reveals that seemingly delayed European approvals were actually processed faster than US approvals when measured in business hours, leading to revised SLA definitions based on regional working hours.
Output
The Shift Activity Time enrichment directly modifies the timestamp values of the selected activity in your event log. No new attributes are created; instead, the existing event timestamps are updated in place. The modified timestamps maintain their original data type and format but with adjusted time values based on your configuration.
The enrichment affects only events matching the specified activity name, leaving all other activities' timestamps unchanged. This selective modification preserves the relative timing between different activities while correcting specific timestamp issues. When "Don't Shift Midnight" is enabled, events with midnight timestamps (00:00:00) remain unmodified, preserving date-only values commonly used in systems that don't track precise times.
After applying this enrichment, all downstream calculations and analyses automatically use the adjusted timestamps. This includes duration calculations, throughput time metrics, bottleneck analysis, and any time-based filtering or segmentation. The changes are permanent within the enriched dataset, so it's recommended to document the time shifts applied for audit trail purposes.
The enrichment handles date boundaries correctly, so shifting timestamps backward may move events to the previous day, while shifting forward may advance them to the next day. Month and year boundaries are also properly handled, ensuring data integrity even with large time shifts.
See Also
- Freeze Time: Set a fixed reference time for the entire dataset, useful for creating consistent snapshots
- Add Days to a Date: Add or subtract days from date attributes when working with date-only values
- Duration Between Two Activities: Calculate time differences between activities after time corrections
- Add Time to attributes: Apply time adjustments to case or event attributes beyond just activity timestamps
This documentation is part of the mindzie Studio process mining platform.