Abstract
{ "background": "The sustainable operation of water treatment infrastructure in developing regions requires robust methods for evaluating system performance. In Rwanda, assessing the yield—the ratio of treated water output to raw water input—of existing facilities is critical for infrastructure planning and investment, yet longitudinal comparative studies are scarce.", "purpose and objectives": "This study aims to methodologically evaluate and compare the operational yield of different water treatment system types in Rwanda using a quasi-experimental design, identifying factors that significantly influence long-term yield improvement.", "methodology": "A comparative quasi-experimental design was employed, analysing longitudinal operational data from matched pairs of conventional rapid sand filtration and dissolved air flotation facilities. Yield was modelled using a fixed-effects panel regression: $Y{it} = \\alpha + \\beta1 X{1,it} + \\beta2 X{2,it} + \\mui + \\epsilon{it}$, where $Y{it}$ is the yield for plant $i$ at time $t$, $X$ denotes explanatory variables (e.g., influent turbidity, maintenance cycles), $\\mui$ represents plant-specific effects, and $\\epsilon{it}$ is the error term. Inference was based on cluster-robust standard errors.", "findings": "The analysis indicates a statistically significant yield advantage for dissolved air flotation systems over the study period, with an estimated mean yield improvement of 12.3 percentage points (95% CI: 9.8, 14.8). This performance gap was most pronounced during high rainfall periods, highlighting the influence of raw water quality variability.", "conclusion": "The quasi-experimental methodology provides a robust framework for isolating the effect of treatment technology on yield from confounding operational factors. The findings demonstrate a clear performance differential between the two dominant system types in the Rwandan context.", "recommendations": "Future infrastructure planning should prioritise technologies demonstrating resilience to source water fluctuations. Implementing continuous yield monitoring integrated with key performance indicators is recommended for all major treatment works