Batch Timestamp Rounding

What This Is

• Batch Timestamp Rounding is the process Factbird uses to normalize batch timestamps to a shared 30-second time grid.
   
• Batch start times (actualStart) are rounded down:
  • :00.000
  • :30.000
     
• Batch stop times (actualStop) are rounded up:
  • :29.999
  • :59.999

Why It Matters

• Consistent timestamp alignment helps Factbird:
  • Calculate OEE accurately
  • Aggregate production data consistently
  • Generate stable KPI reporting
  • Standardize analytics across datasets

When You Would Use This

Use this when:

• Reviewing batch timestamps through the API
• Working with XML batch imports
• Investigating shifted or rounded timestamps
• Building automated batch integrations

How It Works

• Each minute is divided into two 30-second intervals:
  • 00.000 → 29.999
  • 30.000 → 59.999
     
• actualStart values are rounded down to the beginning of the current interval.
  • 11:10:17.500Z → 11:10:00.000Z
     
• actualStop values are rounded up to the end of the current interval.
  • 11:10:17.000Z → 11:10:29.999Z
     
• Because timestamps are normalized after ingestion, batches that occur only seconds apart may appear shifted in stored data.
   
• For XML imports, normalization can also create actual overlaps between consecutive batches.
  • Example:
    • Batch A stop: 11:10:31 → 11:10:59.999
    • Batch B start: 11:10:32 → 11:10:30.000
       
• In this example, the normalized batch ranges overlap even though the original timestamps do not.
   
• Since Factbird does not allow overlapping batches, the second batch may fail to start or import correctly.

Key Terms / Components

• actualStart:
  • The normalized batch start timestamp.
     
• actualStop:
  • The normalized batch stop timestamp.
     
• 30-Second Interval:
  • The fixed time window used for timestamp alignment.
     
• Timestamp Normalization:
  • The process of adjusting timestamps to standard interval boundaries.

Common Misunderstandings

• “The timestamps are incorrect.”
  • The rounding behavior is intentional.
     
• “Gaps between batches mean missing data.”
  • Small gaps or shifts are expected after normalization.
     
• “Exact XML timestamps are preserved.”
  • Imported timestamps are still normalized to 30-second intervals.
     
• “Closely timed batches cannot overlap.”
  • After rounding, batches within the same interval may overlap, which may cause the second batch to fail to start or import.

Related Articles

• When pulling data from the API, why is there a time gap between batches?