The mathematical modeling of a biofilter for effective biological gas removal has been problematic because of uncertain parameters incorporated into complicated equations. These parameters represent the critical factors in biofilter design and operation, such as mass transfer, biodegradation, and subsequent biofilm growth. Such parameters, which are highly site specific, should be estimated in biofilter operation. To resolve this issue, the application of the original Genetic Algorithm (GA) and its modified technique, an improved differential evolution method, was investigated as a global optimization technique, for optimal estimation of the parameters. Previous studies on the application of GA to biofiltration modeling were reviewed. Based on the review, a proper GA application strategy was developed by integrating GA with other conventional optimization and scientific/engineering-based models for enhanced computational efficiency. Thereby, the optimality of the strategy for parameter estimation was confirmed.