An Explainable Forecasting System for Humanitarian Needs Assessment

We present a machine learning system for forecasting forced displacement populations deployed at the Danish Refugee Council (DRC). The system, named Foresight, supports long term forecasts aimed humanitarian response planning. It is explainable, providing evidence and context supporting forecast. Additionally, it scenarios, whereby analysts are able to generate under alternative conditions. has been in deployment since early 2020 powers several downstream business functions within DRC. central our annual Global Displacement Report which informs describe key outcomes, lessons learnt, along with technical limitations challenges deploying systems sector.