<p>We consider an overseas manufacturing supply chain with long lead-times and multiple transportation modes, where orders are placed using forecasted demand. The forecast error, which is the difference between forecasted and actual demand quantity, is considered an uncertain parameter. We assume that the amount of excess inventory in the warehouse at the beginning of each period is also uncertain. Order quantities of parts must be determined for each available transportation mode. We model this problem using a two-stage stochastic programming approach to minimize the overall expected order procurement, inventory holding, and backorder costs under demand and inventory uncertainty. We use the Sample Average Approximation (SAA) method to solve our two-stage stochastic program. We run our experiments using simple random sampling as well as the stratified sampling technique called Latin Hypercube Sampling (LHS) to generate random samples from a continuous distribution in each replication of SAA. We compare the results obtained by using simple random sampling and stratified sampling methods. Further, we use a scenario decomposition based method, Progressive Hedging Algorithm (PHA), to solve the extensive form of two-stage stochastic programming problem generated in each replication of SAA. We evaluate the performance of our solution algorithm at different levels of forecast error and inventory uncertainty.</p>