Publications · 2026
Resilient Last-Mile Allocation in Humanitarian Supply Chains: A Reproducible Four-Objective Fuzzy Framework
Applied Sciences, 16(15), 7581, 2026 · doi:10.3390/app16157581 · Open access
Abstract
Disaster frequency is rising, and response-phase planners face a binding constraint: relief capacity is always smaller than need, so directing affected people to relief centres is a triage decision under time pressure. Published optimisation models speak operations-research vocabulary (cost, distance, equity); civil-protection planners speak supply-chain-resilience vocabulary (the 4R framework of Bruneau et al.). We close the gap with a four-objective humanitarian last-mile allocation model whose objectives map one-to-one onto the 4R components—Resourcefulness, Robustness, Rapidity, Redundancy—evaluated by Mamdani fuzzy inference systems and solved by NSGA-II, NRGA, and NSGA-III. Across 540 runs over three problem sizes, the four-objective formulation surfaces independent Robustness and Rapidity axes that the baseline three-objective formulation fuses into one, a visibility gain established by rank-correlation analysis; NSGA-III is faster than NSGA-II at every size (27–37%, mean 32%), with the hypervolume ranking depending on the reference-point scheme. Weight- and rule-perturbation sweeps confirm the recommendations are robust to expert-elicitation noise, flagging the three-rule Rapidity classifier as a sensitivity hot-spot. The framework is released as a production-grade, openly licensed decision-support tool—continuously tested, dependency-audited, and reproducible bit-for-bit on stock hardware—treating reproducibility as a first-class resilience criterion and positioning audit-ready operational tooling as the digital-technology innovation supply-chain-resilience practice needs.
Abstract reproduced from the published article under CC BY 4.0.
Read the full text at the publisher
Cite
@article{rabiei2026resilient,
title = {{Resilient Last-Mile Allocation in Humanitarian Supply Chains: A Reproducible Four-Objective Fuzzy Framework}},
author = {Rabiei, Peyman and Arias-Aranda, Daniel and Stantchev, Vladimir},
journal = {Applied Sciences},
volume = {16},
number = {15},
pages = {7581},
year = {2026},
publisher = {MDPI AG},
doi = {10.3390/app16157581},
url = {https://doi.org/10.3390/app16157581}
}