search

NAVADMC / ADSM

A simulation of disease spread in livestock populations. Includes detection and containment simulation.
Other
10 stars 5 forks source link

Outputs - Transition State Cumulative #763

Open missyschoenbaum opened 7 years ago

missyschoenbaum commented 7 years ago

This one is somewhat complicated, but I will give it a try and can add on as needed. In initial release, outputs have only daily results. Some of the are easily summarized. Others are not. Transition State is one of those sets that is not easily summarized. Transition State describes how the disease moves through phases, starting at a susceptible state, when infected moving into latent phase, then moving to subclinical that may be transmissible and show some signs, to a clinical state that has disease signs, on to an immune state (either naturally immune or vaccine induced immunity). The model also keeps up with those units that are destroyed, and they could be destroyed for various reasons during any of these states. Each of these disease states is described by a pdf, so the latent phase may be from 1 to 3 days. how to read the variable tsd = transition state daily U/A = Unit/Animal Susc = Susceptible, Lat = Latent, Subc = Subclinical, Clin = Clinical, NImm = Naturally Immune, VImm = Vaccine Immune, Dest = Destroyed In the data, in Results_dailybyproductiontype provides these variables and seems to be reporting how many are in a given state on a given day.
To report a summary, I need to report a distinct number that moved into the state. I cannot summarize the data to find this, because (due to the random pdf value) a unit may sit in the latent state for X days before moving forward. With latent, I could possible do math as my susceptible counts are reduced, but in any of the other state the model may have intervened and selected to test (therefore detect) and destroy units at any of the other states.

I could report a total Unit days in state, but that is not the same. Use this query to see how transition state works. Select iteration, day, last_day, tsdUSusc, tsdULat, tsdUSubc, tsdUClin, tsdUNImm, tsdUVImm, tsdUDest FROM Results_dailybyproductiontype r WHERE 1=1 AND iteration = 1 AND production_type_id is null order by 2 Does something come back from the C-engine that we can assemble this dataset? This would allow me to reproduce the output table OutIterationByProductionType

missyschoenbaum commented 7 years ago

@BryanHurst i am not sure how much of the old app you have seen. I can pull it up and show you if needed. Note this example shows Unit only.