AI-Driven Optimization of Humanitarian Supply Chains for Efficient Disaster Management: Research Gaps and Proposed Methodological Framework
Contributors
Shashi Kant Gupta
Keywords
Proceeding
Track
General Track
License
Copyright (c) 2026 Sustainable Global Societies Initiative

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Abstract
Disruptions to Infrastructures like transportation networks, communication networks, and essential facilities are frequently affected by natural disasters. In the deficiency of these facilities, humanitarian supply chains (HSCs) plays a crucial role to make sure that necessary relief items like food water medicines shelter materials are provided to victims within stipulated time. Though, humanitarian relief logistics is influenced by incomplete information, uncertain demand, limited resources, varying accessibility, and coordination inefficiencies. Such fluctuating situations greatly hinder effective responses by the use of traditional supply logistics. Potential for improvement to make decisions in prediction and situational awareness by Artificial Intelligence (AI) techniques. Moreover, humanitarian operations like facility location, inventory allocation, relief-item prepositioning, transportation planning, and last mile distribution can be reinforced by mathematical optimization techniques. An integrated AI-driven optimization framework for efficient humanitarian supply-chain management is proposed in this paper, in which disaster-related and real-time data, AI-based prediction, uncertainty estimation, multi-objective optimization, and a decision-support system are integrated. The proposed model is intended to minimize unmet demand, Response time, and transportation cost while resource utilization, service coverage, robustness and fairness are pursued to be maximized. Major knowledge gaps related to real-time AI integration, dynamic routing, explainability, multi-disaster modelling, humanitarian equity, communication disruption, and real-world validation are identified. A research methodology involving data collection, preprocessing, AI-based prediction, optimization, simulation, and comparative evaluation is proposed. Through the proposed framework, faster, more adaptive, transparent, and resilient disaster-response decisions are expected to be supported.