{"id":"arxiv/2609.17373","url":"https://arxiv.org/abs/2609.17373","tags":["math.OC"],"score":0,"title":"Data-Driven Stress Testing of Intermodal Freight Networks Using GAN-Generated Disruption Scenarios","domain":"math.OC","source":"arxiv","authors":["Xudong Wang","Mustafa Can Camur","Sabarna Choudhuri","Xueping Li"],"reviews":0,"subject":"math.OC","summary":"Intermodal freight networks are increasingly exposed to correlated, multi-mode disruptions, yet resilience assessments often rely on historical or uncorrelated scenarios that understate systemic risk. This paper develops a data-driven stress-testing framework integrating generative adversarial networks (GANs) with an intermodal optimization model to evaluate performance under realistic compound disruptions. The case study examines weather-related disruptions in the Tennessee Valley corridor. Each GAN-generated scenario is used as a simulation input, and the resulting routing problem is…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-09-15","reactions":{}}