Computer Science Location: Zoom (contact presenter for link)
Add to Calendar 2023-12-08T10:30:00 2023-12-08T10:30:00 America/New_York Ph.D. Qualifying Exam Presentation by Zakaria Mehrab A Data-Driven Theory-Guided Agent-Based Framework to Study Conflict-Induced Forced Migration   Abstract: Zoom (contact presenter for link)

A Data-Driven Theory-Guided Agent-Based Framework to Study Conflict-Induced Forced Migration



Large-scale population displacements arising from conflict-induced forced migration generate uncertainty and introduce several policy challenges. Addressing these concerns requires an interdisciplinary approach that integrates knowledge from both computational modeling and social sciences. Existing migration modeling frameworks that attempt to address policy implications primarily focus on the destination while leaving absent a generalized computational framework grounded by social theory focused on the conflict-induced region. To model conflict-induced migration, we propose a generalized computational agent-based modeling framework grounded by \emph{Theory of Planned Behavior} where we decouple the perception and action phase by utilizing a discounting utility model and a threshold model over different hierarchies of the agent. Our model captures the migration dynamics of agents at fine temporal and spatial granularity. This approach significantly outperforms the State-of-the-Art in capturing the daily temporal trend and scale of migrant outflow for the case of the Russian invasion of Ukraine, achieving a Pearson Correlation Coefficient (PCC) of 0.98 and Root Mean Squared Percentage Error (RMSPE) of 0.14. We also showcase the generalizability of the model by simulating two past conflicts in Burundi and Northern Mali. Sensitivity analysis of the model reveals parameters related to discounting past events and peer influence as significant parameters of the model. We conclude by demonstrating the policy implications of the proposed framework in the form of two case studies. The first case study models the migration behavior of Ukrainian civilians attempting to flee from regions encircled by Russian forces. In another, we estimate the migration pattern under counterfactual scenarios of various event intensities.



  • Anil Vullikanti, Committee Chair (CS, Biocomplexity /SEAS/UVA)
  • Madhav Marathe, Advisor (CS, Biocomplexity /SEAS/UVA)
  • Jundong Li (CS, ECE/SEAS, SDS/UVA )
  • Tariq Iqbal (SYS, CS/SEAS/UVA )