🤖 AI Summary
Refugee geographic allocation often optimizes for aggregate employment rates, inadvertently exacerbating employment disparities across demographic subgroups (e.g., nationality, age, education level).
Method: This paper introduces the first dynamic allocation framework incorporating explicit group-fairness constraints. It integrates machine learning–based individual employment probability prediction with fairness-aware integer programming and online matching algorithms, enabling scalable, multi-round sequential decision-making.
Contribution/Results: Evaluated on real-world humanitarian sector data, the approach achieves up to a 50% improvement in average employment rate over current practices, while reducing inter-subgroup employment rate disparities by over 40%. The framework thus simultaneously enhances global utility and subgroup-level stability, significantly improving both fairness and robustness of allocation outcomes.
📝 Abstract
Ensuring that refugees and asylum seekers thrive (e.g., find employment) in their host countries is a profound humanitarian goal, and a primary driver of employment is the geographic location to which the refugee or asylum seeker is assigned. In the past few years, innovations in analytics have given rise to machine learning (ML) models that predict integration outcomes using personal characteristics. With these ML models, recent research has proposed and implemented algorithms that assign refugees and asylum seekers to geographic locations in a manner that maximizes the average employment. While these algorithms can have substantial overall positive impact (up to 50% increases in average employment rate compared with current practice), using data from two industry collaborators we show that the impact of these algorithms can vary widely across key subgroups based on country of origin, age, or educational background.
