Forecasting human mobility in contexts of crises

The co-hosted IOM-FFO Workshop on Forecasting Human Mobility in Contexts of Crises convened 67 key actors from academia, research, the humanitarian sector, policy and government involved in developing predictive analytics. The aim was to assess the current state-of-the-art in forecasting approaches regarding human mobility in the context of crises. In total 35 different organizations were present for the three day workshop from 22 to 24 October 2019. The workshops sought to address the lack of shared standards for employing automated and model based analytics to inform decisions on where, when and how to respond and assist crisis displaced populations.

This report offers a summary of the discussions held, challenges and best practices identified, and ideas for potential solutions. The introduction lays out the rationale for convening this interdisciplinary workshop. The following sections (II.-IV.) cover the workshop objectives, the rationale and the process of predictive analytics, including its opportunities and limitations. Section V on ethical concerns and possible solutions builds on group work and brainstorming sessions during the workshop. Sections VI. and VII. on best practices outline what has already been done (for example OCHA’s Peer Review work stream2) and where possibilities exist to further build upon existing efforts. The final sections VIII. and IX. cover community needs, namely requirements to improve practices for forecasting human mobility in contexts of crises and to map a potential way forward. The content is not attributed to specific participants since discussions were held under Chatham House Rules.