An optimization approach for brass casting blending problem under aletory and epistemic uncertainties

A critical process in brass casting is the determination of the materials and their quantities to be added into the blend. The reason of being critical is the uncertainty about metal percentages in scrap raw materials. In this paper, the aleatory and epistemic uncertainties, which are modeled by using probability and possibility theory, respectively, have been handled simultaneously in a blending optimization problem for brass casting and a solution approach that transforms the possibilistic uncertainties into probabilistic ones is proposed. A numerical example is performed by the data supplied from MKE brass factory in Turkey. The results of the example have showed that the proposed approach can be effectively used for solving blending problem including aleatory and epistemic uncertainties in brass casting and other scrap based production process. ► We consider probabilistic and possibilistic uncertainties in a blending problem. ► Possibilistic parameters are transformed into the probabilistic ones. ► After the transformation the problem is modeled as a stochastic programming model. ► Finally, stochastic model is solved by using GAMS/Baron solver.