You are here:
MULTIOBJECTIVE PARALLEL GENETIC ALGORITHM FOR WASTE MINIMIZATION
Citation:
Kim**, K. J. AND R L. Smith*. MULTIOBJECTIVE PARALLEL GENETIC ALGORITHM FOR WASTE MINIMIZATION. Presented at AIChE, San Francisco, CA, November 16 - 21, 2003.
Impact/Purpose:
To inform the public
Description:
In this research we have developed an efficient multiobjective parallel genetic algorithm (MOPGA) for waste minimization problems. This MOPGA integrates PGAPack (Levine, 1996) and NSGA-II (Deb, 2000) with novel modifications. PGAPack is a master-slave parallel implementation of a single objective genetic algorithm while NSGA-II is a multiobjective genetic algorithm based on non-dominated sorting of the generated solutions. MOPGA has two main features: (a) more uniform Pareto optimal solutions by modifying the non-dominated sorting module and (b) less function evaluations due to a steady-state genetic algorithm used in PGAPack.
We have applied this efficient MOPGA for separating valuable solvent from waste solvent mixtures using batch distillation. The objectives are to maximize distillation product (i.e., separated solvent), minimize entrainer feeding, and minimize potential environmental impacts (Young and Cabezas, 1999) under given purity constraints. The MOPGA model provides a very linear speedup and high efficiency up to 10 processors used in this research, and provides uniform Pareto optimal solutions that satisfy better economic and environmental performance.
Reference
Deb, K., Agrawal, S., Pratap, A., and Meyarivan, T. (2002) "A Fast Elitist Non-Dominated Sorting Genetic Algorithm for Multi-Objective Optimization: NSGA-II" IEEE Transactions on Evolutionary Computation, 6(2):182-197.
Levine, D (1996) "Users Guide to the PGAPack Parallel Genetic Algorithm Library", Technical Report ANL-95/18.
Young, D.M. and Cabezas, H. (1999) "Designing Sustainable Processes with Simulation: The Waste Reduction (WAR) Algorithm" Comput. Chem. Engng. 23:1477-1491.