SPIN: Pastoralist insecurity forecast model

The SPIN model is a machine learning model that has been developed to predict risk levels in the Sahel countries' pastoralist transhumance corridors on the basis of historical security incidents.

Supporting pastoralist communities with early warning information

The SPIN model was developed in collaboration between Réseau Bilatéral Maroobé (RBM) and DRC with funding from ECHO. The aim was to develop a tool to:

Guide livestock in an environment characterized by drought and insecurity
Identify safe grazing areas using model predictions
Determining the best route for pastoralists and their herds

These objectives aim to provide pastoralists with better visibility of risks, thereby enabling them to anticipate and manage these security challenges effectively.

The SPIN model is built around data collected by the RBM. RBM collects information on alerts through its network of sentinels. These events may include armed attacks, kidnappings, illegal tax levies. This data is combined with data on conflicts linked to events involving livestock or breeders reported by ACLED. Furthermore, data on market prices, as well as natural disasters and environmental conditions are included in the model.

The model is trained on the historical data on these different indicators and use that to provide the forecast of alerts in the coming months in the different regions covered