Generative-AI Decision-Support and Optimisation Framework for Sustainable Logistics Resilience in Industrial Supply Chains – Example of Special-Purpose Mining Shaft Hoist Ropes

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Offers a novel integration of generative AI, multi-objective optimization, and EU policy variables into logistics sustainability research.

🏢実務担当者:Provides a decision-support tool to assess cost-carbon-compliance trade-offs in global sourcing and transport planning under EU rules.

🏛政策担当者:Demonstrates how AI systems can operationalize regulatory variables, suggesting a model for embedding policy into corporate decision tools.

This paper develops a generative AI decision-support and optimisation framework for advancing sustainability and resilience in industrial logistics. The framework combines data aggregation, generative scenario creation, simulation-based evaluation, and multi-objective optimisation to support evidence-based management under tightening European Union sustainability regulations. Building upon the decision-aid lineage of the International Journal of Production Research, it integrates policy variables such as the Carbon Border Adjustment Mechanism (CBAM), the EU Emissions Trading System for maritime transport, FuelEU Maritime, the Digital Product Passport (DPP), and the Corporate Sustainability Reporting Directive (CSRD) directly into logistics-planning equations. Recent studies on digital twins and adaptive optimisation (Longo et al., 2023; Flores-García et al., 2025) highlight the need for AI systems that translate these policies into dynamic cost and carbon trade-offs. The proposed model responds to this need by coupling generative scenario synthesis with traceable optimisation and governance controls consistent with the EU AI Act (European Commission, 2025). An illustrative case from the mining-rope industry demonstrates how global sourcing and transport routes in European, South African, and Chinese configurations can be simulated within the generative environment to evaluate comparative cost, emission, and compliance profiles. Both SME-light and enterprise implementations achieved reduced analysis time and improved transparency of carbon-related decisions. The study contributes a replicable methodology that transforms generative AI from a creative text tool into a quantifiable governance instrument, linking strategic foresight with operational resilience in sustainable logistics networks.

• openalex https://doi.org/10.20944/preprints202601.0628.v1first seen 2026-08-02 19:28:07