Case calculation and loading

The data load process involves sequentially considering each anchor object and calculating cases up to a predefined case limit. The approach ensures that the case allocation bucket is filled before any deletions occur, adhering to the specified purge type.

This is the case calculation methodology:

  • Single anchor object
    • For processes with one anchor object, the data load calculates cases based on the anchor up to the case limit for the specified date range.
  • Multiple anchor objects
    • Two anchor objects
      • Calculate all cases from the first anchor.
      • If the case limit is not reached, calculate cases from the second anchor up to the case limit.
    • Three anchor objects
      • Calculate all cases from the first anchor.
      • If the case limit is not reached, calculate cases from the second anchor.
      • If the case limit is still not reached, calculate cases from the third anchor up to the case limit.
    • Additional anchors
      • Continue the process for any additional anchors in a similar sequential manner.

Example scenario

  • Allocated cases: 350,000
  • Date range: January 1, 2022 - December 31, 2022
  • Anchor objects
    • Purchase order line
    • Requisition line
    • Claim line

Data load process

  • Purchase order line
    • Calculate cases between January 1, 2022, and December 31, 2022, which have purchase order line.
    • If there are more than 350,000 cases, keep only 350,000 and purge the rest based on the selected type of purge.
    • If there are fewer than 350,000 cases (for example, 200,000), proceed to the next anchor.
  • Requisition line
    • Calculate cases between January 1, 2022, and December 31, 2022, which have requisition line.
    • If there are more than 150,000 cases, keep only 150,000 and purge the rest based on the selected type of purge.
    • If there are fewer than 150,000 cases (for example, 100,000), proceed to the next anchor.
  • Claim line
    • Calculate cases between January 1, 2022, and December 31, 2022, which have claim line.
    • If there are more than 50,000 cases, keep only 50,000 and purge the rest based on the selected type of purge.
    • If there are fewer than 50,000 cases (for example., 20,000), the remaining 30,000 cases are available.

Important considerations

  • Initial events: The data load is based on the initial events captured within the specified date range.
  • Data completeness: as the anchors are executed based on priority, it might be that only the 2nd or 3rd anchor hits the case allocation limit. In that instance, the purge type will come into play and remove cases for this specific anchor based on the purge type. This can lead to the scenario where a PO line (1st anchor) covers the whole date range (Jan 1st – Dec 31st ), whereas the requisition line (2nd anchor) only covers cases up to the allocation limit, which could be hit in August, so the data for requisition line would be, for example, from Jan 1st – August 15th .
  • Purge type
    • Purge latest data: Keep cases from the start of the date range until a specific day (day X).
    • Purge oldest data: Start adding cases from the end of the date range until a specific day (day Y).