UTLDR is an agent-based, stochastic, frameworks that can be used to simulate the effect of alternative combinations of public interventions.
UTLDR name is an acronym of the "meta-compartments" that compose it: Undetected - Tested - Lockdown - Dead - Recovered.
It requires as input a population of agents (preferibly characterized by age/gender as well as working place) tied in a social tissue (it is not a mean field approach). Household structures as well as schools, workingplaces and other social circles are assumend to be embedded within the social network and annotated at agent level.
- UTLDR jupyter notebook with incremental examples of the base model
- UTLDR2 source code: extension of UTLDR that leverages geographical tesselation (at census cell level)