The post-processing pipeline is used to process all results, creating useful maps and graphics. Currently, the following maps are generated for the base run and each control type:
- For infections that were originally seeded in the given county:
- Number of animals infected
- Number of premises infected
- Epidemic Extent (number of counties infected)
- Duration of Infection
- County Risk (number of times the given county was infected given that it was not the seeded county)
- Proportion of infections attributed to local spread
- Proportion of infections attributed to shipment spread
- Most common routes of infection
- Least common routes of infection
Along with the maps, the post-processing code generates a variety of tables and graphs. A general list of these is as follows:
- Table of summary statistics to compare control types.
- QQplots comparing the distributions of each control type.
- Violin plots for exploratory analysis of the effect of control types.
- Regression and graphics to determine the effect of premise density, clustering, in-shipment number, and out-shipment number on the effectiveness of control.