A robust optimization model for humanitarian relief chain design under uncertainty

Due to the importance of relief operations in disasters, this paper aims to contribute humanitarian logistics under uncertainty. In this paper, a three-level relief chain model consisting of suppliers, relief distribution centers, and affected areas is considered. The uncertainty associated with demand, supply, and all of the cost parameters is addressed by employing robust optimization, where the uncertain parameters are independent and bounded random variables. While the proposed model attempts to minimize the total costs of the relief chain, it implicitly maximizes people's satisfaction level in the affected areas through applying a penalty to shortages of relief commodities. Additionally, a data set derived from a real disaster case study in the Alborz area, which is vulnerable to earthquakes, is applied to test the efficiency of the proposed robust relief chain model compared to its deterministic form. The study analyzes the degree to which each uncertain parameter affects the solution of the relief chain model and consequently helps the decision maker to tune the parameter values more accurately.