Logistic networks for product recovery have to be implemented in an efficient manner to recover cost savings from remanufacturing. Uncertainty of the quantity and timing of product returns is a major issue in the product recovery networks. Product recovery networks require investments of high fixed costs. Hence the uncertain information has to be taken into account when the strategic network model is designed. This thesis is aimed at presenting a mathematical model for closed-loop product recovery system with uncertainty of product returns. A generic mixed integer-programming model is developed. Stochastic programming approach is implemented in the deterministic model by adding scenarios and probabilities to explicitly account for the uncertainties in the product returns. The model is programmed and solved by LINGO optimization solver. Several test problems are solved by varying the parameters: number of scenarios, probability and return rates to identify the sensitivity of the model. A statistical analysis is conducted on all the example problems by measuring the Expected Value of Perfect Information (EVPI), Value of Stochastic Solution (VSS) and the results are discussed.