The stochastic transportation problem with single sourcing

We propose a branch-and-price algorithm for solving a class of stochastic transportation problems with single-sourcing constraints. Our approach allows for general demand distributions, nonlinear cost structures, and capacity expansion opportunities. The pricing problem is a knapsack problem with variable item sizes and concave costs that is interesting in its own right. We perform an extensive set of computational experiments illustrating the efficacy of our approach. In addition, we study the cost of the single-sourcing constraints.

► We study a single-period stochastic transportation problem.
► Deliveries are of a single-sourcing nature and cost structures are nonlinear.
► We develop an effective branch-and-price algorithm for solving the problem.
► The algorithm efficiently solves a wide range of test problems.