Capacity management of migrant accommodation centers using approximate dynamic programming

Irregular migration has become a major macro-economic and political challenge. Due to rising conflicts income inequality across the world, number of migrants is expected increase exponentially over coming decade. Thus, it critical importance effectively use limited resources allocated humanitarian operations for irregular migration. In this paper we model problem capacity management migrant transfers within network accommodation centres with stochastic dynamic programming. Our study extends literature on modelling by applying Approximate Dynamic Programming (ADP) into new context. The translatable in other similar migratory routes locations around world where governments need deal uncertain numbers migrants. We test our approach five Greek islands which have been main arrival points during European Migrant Crisis. results show that ADP provides better computational performance than simple myopic heuristic. sensitivity analysis gives insights decision-makers about impact parameter values policies.