Priority-Driven Dynamic Constraint Boundary Relaxation: A Framework for Multi-Constraint Conflict Resolution
IEEE Access · 2026 · 14 · 126311-126319
DOI: 10.1109/access.2026.3723836 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Yang Liu. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This paper presents a priority-based dynamic constraint boundary relaxation (PDCBR) framework for managing multi-constraint conflicts. In the first stage, a two-dimensional model incorporating rigidity and impact intensity sets the constraint-satisfaction order and classifies constraints as core, buffer, or peripheral. The second stage uses conflict intensity to apply differentiated boundary adjustments in normalized-slack space; this construction is invariant to sign and applies uniformly to upper-bound, lower-bound, equality, and two-sided constraints. Core boundaries remain unchanged by design, so every solution feasible for the relaxed problem satisfies the original core constraints exactly, irrespective of the solver. The study establishes conditional links to penalty formulations: an asymptotic multiplier correspondence for quadratic penalties and a finite-parameter correspondence for exact L1 penalties under standard constraint qualifications. It also shows that relaxation leaves the objective value of any fixed solution unchanged, while the global Pareto front may expand when new feasible points appear; both effects are measured against independently enumerated reference fronts. Evidence comes from three seeded and reproducible case studies (multi-constraint knapsack, project scheduling, and commercial bank resource allocation), a 324-configuration exact sensitivity analysis, scalability tests with multi-restart local search, and validation on OR-Library benchmark instances. Results are exact under full enumeration and are otherwise explicitly identified as local optima; complete code and data are released.
Zitieren
APA
Liu, Y. (2026). Priority-Driven Dynamic Constraint Boundary Relaxation: A Framework for Multi-Constraint Conflict Resolution. IEEE Access, 14, 126311-126319. https://doi.org/10.1109/access.2026.3723836
Vancouver
Liu Y. Priority-Driven Dynamic Constraint Boundary Relaxation: A Framework for Multi-Constraint Conflict Resolution. IEEE Access. 2026;14:126311-126319. doi:10.1109/access.2026.3723836
BibTeX
@article{Liu2026,
author = {Liu, Yang},
title = {Priority-Driven Dynamic Constraint Boundary Relaxation: A Framework for Multi-Constraint Conflict Resolution},
journal = {IEEE Access},
year = {2026},
volume = {14},
pages = {126311--126319},
doi = {10.1109/access.2026.3723836}
}
EUVIMED