Today’s supply chains are more and more complex. They depend on a network of independent, yet interconnected moving parts. They rely on critical infrastructures and experience a lot of time variability and randomness. Designing strategies that deal with such constantly changing supply chains is necessary in this increasingly globalized economy where supply chain disruptions have impacts that propagate not only locally but also globally. In this paper we propose a randomized flow control algorithm for a time varying, random supply chain network. We formulate a constrained stochastic optimization problem that maximizes the profit function in terms of the long-run, time-average rates of the flows in the supply chain. We show that our algorithm, which is based on queueing theory and stochastic analysis concepts, can get arbitrarily close to the solution of the aforementioned optimization problem. In addition, we describe how the flow control algorithm can be extended to a multiple firms supply chain setup and present numerical simulations of our algorithm for different supply chain topologies.