This article considers how best to strategically transport and route hazardous materials through a bi-modal transportation network (MMTS) consisting of road and rail links. The placement of transfer yards is also considered as well as the possibility of disruptions at those facilities. For this decision problem a two-stage stochastic programming model is formulated. The objective of the model is to minimize transportation costs and risk. As the model is intractable to solve using traditional techniques, three algorithms are proposed, namely Maximum Likelihood Sampling (MLS), Sample Average Approximation (SAA), and a combination of these two algorithms (MLSAA). Those optimization approaches have been applied to various instances to demonstrate their viability and effectiveness. Numerical testing shows that the MLSAA algorithm is superior and can solve large-scale instances.
