Bug - activity/infected time handling in Calculator
The infected person presence time is not handled correctly. I think it's due to the way we defined the constraints last week, where we didn't case so much about the case of the infected person falling outside of the activity period -
Case 1: Infected person present from 09:00 to 13:00, exposed person present from 14:00 to 18:00. P(i) = 14% (this is the result if they are in the space simultaneously) Case 2: Infected person present from 14:00 to 18:00, exposed person present from 09:00 to 13:00. P(i) = 14% (expecting 0!) Case 3: Infected person present from 09:00 to 10:00, exposed person present from 09:00 to 18:00. P(i) = 2% (I'd be expecting a bit more due to the removal time for the ?) Case 4: Infected person present from 09:00 to 10:00, exposed person present from 09:00 to 10:00. P(i) = 2%
The graphs are also identical for these cases. PDF's for each case are attached.
It seems like
- We're only running the model for the infected person present time period (if < activity duration)
- As a consequence we're only displaying and integrating the infected person time.
I think in order to fix this we need to ensure something like: Simulation starts at the minimum of (Activity Start, Infected person present) Simulation ends at the maximum of (Activity Finish, Infected person leaves)
I don't know if this falls in the UI for calling the model @gaazzopa or is due to the way the model is handling the inputs @nimounetexposure_time_bug.pdf