The current COVID-19 pandemic has highlighted the need for data and the value of models for response strategies. In response to this, the John Hopkins University Applied Physics Laboratory (APL) and the Centre for Humanitarian Data developed a model fit for use in humanitarian contexts: OCHA-Bucky. This model can be used to predict the number of cases, hospitalisations, and deaths over the next four weeks. The tool aims to help decision making on a short-term and operational basis, thereby protecting and saving people amidst this humanitarian crisis.
This project began in March, and up until now researchers have identified a number of key take-aways:
- Short-term, regularly updated projections are most helpful.
- Projections should take challenges and risk-factors characteristic of humanitarian crises into account.
- Subnational projections are vital in order to inform operational decisions.
- Decision makers benefit greatly from a simulation tool.
- Results are best presented in a practical way.
OCHA-Bucky is currently available in several countries, including Afghanistan, the Democratic Republic of Congo (DRC), Iraq, Somalia, South Sudan and Sudan. The code, as well as the documentation used, is freely available for anyone in an effort share knowledge and improve the model for other uses or locations. Read more about the project through their blog.