Humanitarian organizations (HOs) often face significant operational challenges during sudden-onset disasters, including constraints on staff, storage, and transportation capacities, which can delay the distribution of critical donations to people in need (PIN). These inefficiencies, often overlooked in the literature, create a trade-off: centralized operations by HOs facilitate accurate demand estimation but may be hindered by bottlenecks, while decentralized donations---where individual donors distribute supplies based on their preferences---offer faster response times but risk supply-demand mismatches.

To address this, we propose a model that partially decentralizes humanitarian operations while accounting for random demand. In the deterministic case, we demonstrate that determining the optimal decentralization level reduces to a one-dimensional piecewise-linear convex optimization problem, enabling a novel search algorithm to identify all optimal solutions. We further prove that the objective function comprises at most 3n+2 pieces, allowing the development of an improved O(n)-complexity algorithm for n affected zones. These results extend naturally to the stochastic demand case, accommodating |Ω| demand scenarios with a complexity of O(|Ω| n).

Moreover, we derive closed-form expressions characterizing the optimal decentralization level, enabling detailed comparative analyses of parameter changes. Our model can also integrate demand uncertainty, donor preferences, and fairness considerations, with these extensions solvable efficiently using the proposed algorithms.

As a case study, we apply our framework to the early stages of humanitarian aid distribution in Ukraine during the Russo-Ukrainian War in 2022. Our findings reveal that when the efficiency discount of HOs is relatively high, partial decentralization proves nearly three times more effective than full centralization, where all donations are routed by trucks across borders. Furthermore, a 33% decentralization level can reduce demand shortage costs by up to 80% compared to full decentralization via postal services.

Lin Yinuo is a fifth-year PhD candidate in the Department of Industrial Systems Engineering and Management at the National University of Singapore, advised by Dr. Li Xiaobo. He received his Bachelor’s degree in Computer Science and Technology from Sun Yat-sen University. His research interests include supply chain, revenue management, robust optimization and the design of algorithms for efficiently solving optimization problems.