Management of environmental pollution control problems under stochastic uncertainty

This study investigated the applicability of using genetic algorithm for tackling chance-constrained programming models utilized for solving environmental pollution-control management problems. Compared with conventional chance-constrained methods, the proposed one could deal with stochastic models with both the left- and right-hand-side constraints being involved with random variables. Two study cases which were related to air quality management and river water pollution control were applied to illustrate the applicability of the proposed method. Both study cases had needs of seeking cost-effective management schemes under uncertainty. The study results indicated that the GA-based CCP models could effectively communicate uncertainties into optimization process, and generate solutions that contain a spectrum of potential waste treatment options with both risk and cost information. Decision alternatives could be obtained by analyzing tradeoffs between the waste handling cost and the system-failure risk due to inherent uncertainties. Compared with the linearization solution method of traditional CCP models, the introduction a GA algorithm could facilitate the solution of more general stochastic models. The proposed method is not restricted to environmental problems and can also be applied to many other engineering management systems.