Multi-objective Optimization Approach for Robust Design under Uncertainty

This paper presents a new strategy for synthesizing an appropriate process with multi-objective under uncertainty. The uncertainty is classified depending on its sources and mathematical model structure as deterministic or stochastic. The proposed methodology is a two-layer algorithm. In the outer layer, the synthesis problem is represented by a multi-objective optimization problem considering the performances associated with design parameters. In the inner layer, the problem is expressed as a singleobjective optimization problem taking in to account the operating performances in the presence of uncertainty. The formulated design problem is solved to construct a trade-off set. A set of feasible solution is then selected through analyzing the constructed trade-off set. The proposed methodology is implemented by integrating in-house software with commercial software tools. We illustrate the applicability of the proposed methodology in a designing of an ethanol dehydration process. Two main technological schemes, azeotropic distillation and pervaporation distillation hybrid processes, and a newly developed reverse osmosis membrane process are investigated to compare their economic, environmental performances under uncertainty. The developed methodology can select solutions with minimal environmental impacts and adequate flexibility at a desired economic performance.