The increased awareness of environmental sustainability has led to increasing attention to closed loop supply chains (CLSC). The main objective of the CLSC is to capture values from end-of-life (EOL) products in a way that ensures a business to be economically and environmentally sustainable. The challenge is the complexity that occurrs due to closing the loop. At the same time, considering stochastic variables will increase the realism of the obtained results as well as the complexity of the model. This study aims to design a CLSC for durable products using a multistage stochastic model in mixed-integer linear programming (MILP) while considering uncertainty in demand, return rate, and return quality. Demand was described by a normal distribution whereas return rate and return quality were represented by a set of discrete possible outcomes with a specific probability. The objective function was to maximize the profit in a multi-period and multi-echelon CLSC. The multistage stochastic model was tested on a real case study at an air-conditioning company. The computational results identified which facilities should be opened in the reversed loop to optimize profit. The results showed that the CLSC resulted in a reduction in purchasing costs by 52%, an annual savings of 831,150 USD, and extra annual revenue of 5459 USD from selling raw material at a material market. However, the transportation cost increased by an additional annual cost of 6457 USD, and the various recovery processes costs were annually about 152,897 USD. By running the model for nine years, the breakeven point will be after three years of establishing the CLSC and after the annual profit increases by 1.92%. In conclusion, the results of this research provide valuable analysis that may support decision-makers in supply chain planning regarding the feasibility of converting the forward chain to closed loop supply chain for durable products.