The Application of Statistical ANOVA, LSD and RSM to Agro-Based Filter Design Optimization

Filtration is one of the practical technologies for water and wastewater treatment. Yet, such a technique involves various factors considered in the design, including capital costs, electricity consumption, labour force, and technical operations to ensure optimal performance. The process of filtration process has become economically and environmentally feasible with the emergence of agro-based filter media. However, more work is still required to ensure successful implementation. This paper evaluates filter design efficiency using analysis of variance (ANOVA) and least significant difference (LSD) test. It proposes a framework to optimize the agro-based filter design with response surface methodology (RSM); by utilizing secondary data from historical works. The findings show that both responses in Case 1 are significant (p ≤ 0.05) in TSS and turbidity removal. Then the filter × 8 in Case 2 produced the highest removal efficiency (TSS = 39.61 ± 1.64; turbidity = 32.55 ± 2.84). As for Case 3, a 3-D surface plot was plotted for three response variables: configuration, loading rate, and turbidity. Hence, it suggests the potential of the agro-based filter as a sustainable alternative.