Solving Real-World Multi-Objective Problems Using Aetheric-Flux Optimization (AFO): A Supply Chain Case Study

Introduction: Multi-objective optimization is essential in real-world systems where multiple conflicting goals must be satisfied simultaneously. In supply chain management, decision-makers aim to minimize operational cost and delivery delay while maximizing service reliability. However, dynamic demand patterns, traffic variability, weather disruptions, and operational uncertainties make real-world optimization more complex than benchmark problems. Many existing algorithms suffer from premature convergence and loss of diversity, leading to unstable Pareto solutions.Objective: This study aims to evaluate the effectiveness of Aetheric-Flux Optimization (AFO), a hybrid metaheuristic algorithm, for solving a real-world multi-objective supply chain optimization problem under dynamic conditions.Method: AFO integrates ACO-style exploration with WOA-style exploitation and incorporates a Dynamic Flux Controller to balance search phases adaptively. An Aether Memory Layer stores and reuses high-quality past solutions to prevent stagnation. The algorithm is tested on a time-indexed logistics dataset containing 32,065 records and 26 features (2021–2024). Three objectives are considered: minimizing cost, minimizing delay, and maximizing reliability. Performance is compared with NSGA-II, MOPSO, GWO, and WOA.Results: AFO achieves superior convergence and diversity (IGD = 0.024 ± 0.005, HV = 0.495 ± 0.010, Spread = 0.18 ± 0.03). It provides a 9.5%, cost reduction, 5.9-hour average delay, and 0.90 reliability. Wilcoxon tests confirm statistical significance (p < 0.05).Conclusion: AFO demonstrates improved stability and practical effectiveness for dynamic multi-objective supply chain optimization problems.