Robust Optimization of Transportation Networks Using Fuzzy and Interval Methods

Abstract

In this paper, we present a robust optimization framework for transportation net works where key parameters such as transportation costs, supplies, and demands are subject to uncertainty. We model these parameters as fuzzy numbers and intervals and propose a novel splitting algorithm to derive solutions that minimize worst-case regret. Rigorous proofs establish the convexity of the objective function and the con vergence of our algorithm. Extensive numerical experiments and graphical analyses demonstrate that our robust methods yield solutions that remain near-optimal even under severe data uncertainty. Our contributions extend classical transportation mod els to more realistic settings and offer significant advantages for practical supply chain decision-making.