The following plot shows the allocated time for CF production (red) and processing time (blue) for each dataset. The upper horizontal bars represent the best scenario if the calibration team member would start 100% of the time to produce the CF's. The number of calibration files assumed is given in brackets.

Assumptions:

- 1 CF per 1 year of data (360 L0 files)

- Processing duration of 6 minutes for one L0 file assumed based on dagster runtime observations, but can be adjusted

- Total processing duration multiplied by 1.5 since theoretically the processing time is at minimum doubled since the calibration staff is doing the initial processing first to do the quality checks, and triggers the operational processing afterwards. Moreover, multiple validity periods would also need a second, third,... processing iteration for the same L0 file, although the netto processing time will certainly decrease. However, this depends where the next validity period has to be introduced.

- The estimated finished date uses the current starting date as a start and counts working days into the future, but excludes the weekends.

Dataset filtering:

- Default lists all datasets, including non-official.

- Default minimum number of L0 files is 15.

- Lab-locations and Testing facilities (Aldino,GreenbeltMD,ElkridgeMD,ColumbiaMD are excluded, except for the official units (P2,P32,P71).

Usage:

This app must go from the left tab to the right. First, the settings have to be made in the first tab. Second, going in the overview tab creates the time estimation for each datasets. And finally, the estimation tab is used the proper values for visualization.