(189r) Multiobjective Tabu Search for Plant Design Models

Multiobjective optimization techniques are employed to visualize the relationship between operating parameters and efficiencies related to profit, emissions, and safety. The challenge in generating trade-off curves for these attributes comes from the nonlinearity and complexity of plant design models, so higher level optimization heuristics are needed to compute the Pareto optimality curves and surfaces. The purpose of this research is to investigate the efficacy and capabilities of the Tabu Search algorithm for multi-objective optimization specifically for plant design projects. A Tabu Search algorithm was developed to solve the 0-1 knapsack problem for multiple objectives. Multiple plant design optimization problems are then solved to determine Pareto surfaces. Results show how tuning the parameters of the algorithm leads to more efficient determination of the Pareto surface, for highly nonlinear plant design models.