This study characterizes the efficient preliminary-design domain of activated-sludge secondary treatment (ST) in wastewater treatment plants (WWTPs) by integrating geometric, economic, and environmental criteria. A parametric Building Information Modeling (BIM) workflow in Revit-Dynamo was coupled with Generative Design (GD) and the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to minimize direct construction cost and embodied CO2-eq over life-cycle modules A1-A5, while independently maximizing the installed useful volumes of the biological reactor (BR) and secondary clarifier (SC). The analysis varied population equivalent (PE), biochemical oxygen demand (BOD), mixed liquor suspended solids (MLSS), geometry, and modularity, limited to two BRs and two SCs. Three random-seed runs generated 340, 323, and 295 non-dominated solutions. The reference run occupied an observed non-dominated domain of approximately 1800-5800 PE and 89.7-256.9 g BOD/PE·day, comprising a region predominantly occupied by 1 BR configurations, a modular coexistence and transition band, and a region without observed non-dominated solutions under the adopted bounds. Cost and CO2-eq were almost perfectly coupled across seeds, with Spearman’s rank correlation coefficients ranging from 0.9978 to 0.9980; BOD showed a stronger association with CO2-eq/PE than PE; and the BR contributed 88.08-89.64% of ST embodied CO2-eq. The workflow provides a traceable decision domain whose empirical envelopes characterize the explored solution space rather than universal feasibility limits.
Copyrights © 2026