Quantum-inspired NSGA-III-BERT hybrid decision engine for high-frequency supply chain resilience configuration: design, simulation and engineering framework
- 1College of Economics and Management, Hefei Normal University
Abstract
High-frequency supply-chain disruptions require decision methods that combine many-objective search with policy-text constraints, yet hybrid systems are often described without sufficiently precise mathematical and evidentiary boundaries. This study presents a classical-computing Quantum-Inspired NSGA-III-BERT decision engine with three separable components: probabilistic population search, policy-aware dynamic reference-point adjustment, and an edge-cloud deployment layer. The robust formulation fixes binary route-activation decisions in the first stage and restricts affine recourse to continuous flow, inventory, and reserve-capacity variables. Across the reported 30-run summaries, the proposed method obtained HV = 0.947 with SD = 0.012 and IGD = 0.082 with SD = 0.007, together with cost and carbon reductions of 19.7% (SD = 0.7%) and 24.5% (SD = 0.9%), respectively. Because the archived package does not preserve paired run-level outputs, inferential comparisons are reported as summary-statistic Welch tests with Holm correction rather than as reconstructed paired tests. Pilot-enterprise observations are interpreted as descriptive deployment evidence rather than causal estimates. The study therefore contributes an explicit algorithm, a mathematically consistent recourse model, and an evidence-bounded interpretation of the empirical results.
Related
- Feasibility of vault configuration of the Neuroform Atlas stent in the treatment of bifurcation intracranial aneurysm: clinical observation and hemodynamic simulation
- A framework to determine the optimum number and configuration of wastewater treatment plants
- Determinants and typologies of health system resilience: A data-driven framework for prioritising investment
- A preliminar validation-driven experimental–numerical framework for vaginal delivery simulation
- Agency and resilience among young persons in Sweden: Patterns of interrelated resources within the WHO Adolescent Well-being Framework
BibTeX
@article{li2026quantuminspired,
title = {{Quantum-inspired NSGA-III-BERT hybrid decision engine for high-frequency supply chain resilience configuration: design, simulation and engineering framework}},
author = {Wei Li},
journal = {Scientific Reports},
year = {2026},
month = aug,
doi = {10.1038/s41598-026-68917-5},
url = {https://doi.org/10.1038/s41598-026-68917-5},
note = {CC BY},
}
RIS
TY - JOUR
TI - Quantum-inspired NSGA-III-BERT hybrid decision engine for high-frequency supply chain resilience configuration: design, simulation and engineering framework
AU - Wei Li
PY - 2026
DA - 2026/08/01
JO - Scientific Reports
DO - 10.1038/s41598-026-68917-5
UR - https://doi.org/10.1038/s41598-026-68917-5
AB - High-frequency supply-chain disruptions require decision methods that combine many-objective search with policy-text constraints, yet hybrid systems are often described without sufficiently precise mathematical and evidentiary boundaries. This study presents a classical-computing Quantum-Inspired NSGA-III-BERT decision engine with three separable components: probabilistic population search, policy-aware dynamic reference-point adjustment, and an edge-cloud deployment layer. The robust formulation fixes binary route-activation decisions in the first stage and restricts affine recourse to continuous flow, inventory, and reserve-capacity variables. Across the reported 30-run summaries, the proposed method obtained HV = 0.947 with SD = 0.012 and IGD = 0.082 with SD = 0.007, together with cost and carbon reductions of 19.7% (SD = 0.7%) and 24.5% (SD = 0.9%), respectively. Because the archived package does not preserve paired run-level outputs, inferential comparisons are reported as summary-statistic Welch tests with Holm correction rather than as reconstructed paired tests. Pilot-enterprise observations are interpreted as descriptive deployment evidence rather than causal estimates. The study therefore contributes an explicit algorithm, a mathematically consistent recourse model, and an evidence-bounded interpretation of the empirical results.
SN - 2045-2322
ER -