To create frequency based data:

1. Select Ignore from the drop-down list labelled ‘Use/Ignore type IDs’.
2. Click the ‘Copy time intervals ’ button to add the correct time intervals for the Arrival Profile in question.
3. To specify arrival frequencies:

   If you would like the same number of arrivals per interval throughout the simulation, enter the number in the first cell labelled ‘All’.

   or

   Enter specific numbers of arrivals per interval in the appropriate cells along the timeline.

4. To set the minimum time between arrivals:

   If you would like the same minimum time between arrivals across all intervals throughout the simulation, enter the time (in seconds) in the first cell labelled ‘All’.

   or

   Enter specific minimum times (in seconds) in the appropriate cells along the timeline.

5. If you want arrival times to be randomised within the constraints of the time period, select Yes from the drop-down list labelled ‘Make arrivals random?’, otherwise leave as No.
   Note: If you don’t randomise arrivals, the first arrival will occur at the very start of the time interval.
6. To add random noise to the amount of Entities in your Arrival Profile, enter ‘Lower’ and ‘Upper’ limit values in the appropriate cells.

   As an example, if you set lower and upper limits of 2 and 5, a cell containing an initial quantity ‘10’ will ultimately be modified to contain between 5 & 8 (if noise is subtracted) and between 12 and 15 (if noise is added). The demand total is preserved regardless of noise.

7. To spread the arrival of Entities over a certain period, add a value (in seconds) to the cell labelled ‘Spread arrivals (sec)’. This ensures that Entities don’t all arrive in one instantaneous surge, which may be unrealistic.
   Note: This spread will override the spread defined on the Data Profiles worksheet.

This portion of the worksheet should now resemble this screen shot.

After import, these settings will create the same six Arrival Profiles mentioned in Splitting an Origin into Suborigins . However, the data for the suborigins (the train cars) will be divided into 10 arrival ‘spikes’. This is due to the ‘Arrival frequency’ of ‘10’ per period (in this case the period is the whole runtime of one hour). This graph illustrates these spikes after import.