A Conflict-Sensitive Mixed Integer Linear Programming Framework for Humanitarian Aid Distribution in Active Conflict Zones

Abstract

The delivery of humanitarian aid in active conflict zones presents a fundamental operational dilemma: traditional logistics optimization prioritizes cost-efficiency, while conflict sensitivity demands security-aware planning (objectives typically treated as separate rather than integrated concerns). This paper addresses this gap by developing a novel Mixed Integer Linear Programming (MILP) framework that systematically incorporates real-time security data into humanitarian distribution optimization. The model advances beyond static risk proxies by designing a mathematical structure capable of ingesting dynamic conflict event data (e.g., ACLED reports) alongside infrastructure constraints, route-specific delay factors, warehouse capacities, and prioritized demand. The framework operationalizes conflict sensitivity through a multi-objective function that minimizes both tangible transportation costs and intangible security risk exposure, subject to constraints including minimum demand fulfillment (≥80%), supply capacity limits, and maximum acceptable risk thresholds. The model incorporates a calibrated risk-weighting parameter (α) that enables humanitarian planners to explicitly quantify and control the trade-off between operational efficiency and personnel safety. A proof-of-concept validation using data from Northeast Nigeria demonstrates the model's computational tractability and its ability to generate feasible solutions that balance competing objectives. The key methodological contribution lies in the framework's structural readiness for temporally relevant security integration—moving beyond probabilistic planning toward event-driven operational optimization. This research provides humanitarian organizations with a defensible, mathematically rigorous tool for making transparent trade-offs between efficiency, security, and equity in the world's most challenging environments.