Reliability-Based Multi-Objective Optimization of Groundwater Remediation

In-situ bioremediation of groundwater is relatively low cost and has high efficiency in remediating groundwater contaminated with petroleum hydrocarbons under suitable hydrogeologic settings. This work develops a multiobjective simulation-optimization (S-O) model for the design of an in-situ bioremediation system for petroleum-hydrocarbon contaminated groundwater. Minimizing the cost of the remediation system (installation and operation) and maximizing its reliability are the two objectives of the developed S-O model. The BIO PLUME II software simulates the remediation process and the non-dominated sorting genetic algorithm (NSGA) II optimizes remediation. The reliability objective measures the effect of uncertainty in the estimate of the initial contaminant concentration on the performance of bioremediation design, and is evaluated under five scenarios of initial contaminant concentration in an example case study illustrating this paper's methodology. The S-O model for optimal remedation calculates Pareto fronts reflecting the best tradeoff between cost and system reliability that can be obtained. Remediation managers choose remediation strategies from the calculated Pareto front that best serve their cost preferences and remediation requirements. The calculated remediation demonstrates the effectiveness of the remediation system is sensitive to the magnitude of the initial contaminant concentration.