{ "background": "Ensuring reliable water treatment is a critical infrastructure challenge in many developing regions. System reliability assessments often fail to account for the hierarchical structure of facility data, where individual plants are nested within regional management zones, potentially biasing results.", "purpose and objectives": "This study aimed to develop and apply a multilevel regression framework to evaluate the reliability of water treatment systems, explicitly modelling hierarchical data structures to provide more accurate performance diagnostics and identify key influencing factors.", "methodology": "We employed a two-level hierarchical linear model. Data from a sample of treatment facilities were analysed, with facility-level measurements (Level 1) nested within administrative regions (Level 2). The core reliability model is specified as $\\text{Reliability}{ij} = \\beta{0j} + \\beta{1}X{ij} + e{ij}$, where $\\beta{0j} = \\gamma{00} + \\gamma{01}Zj + u{0j}$. Robust standard errors were used for inference.", "findings": "The multilevel model revealed significant regional variation in baseline reliability (intercept variance = 0.18, p < 0.01), which a standard ordinary least squares regression failed to detect. A key concrete result is that facilities with automated residual chlorine monitoring were associated with a 15.2 percentage-point increase in reliability scores (95% CI: 11.4 to 19.0).", "conclusion": "The analysis confirms that accounting for data hierarchy is methodologically essential for accurate performance assessment. Regional management practices constitute a significant source of reliability variation, overshadowing some facility-level factors in conventional models.", "recommendations": "Resource allocation and policy interventions should be informed by multilevel analyses to target both regional governance and specific facility-level upgrades. Regulatory frameworks should mandate the reporting of data suitable for hierarchical analysis.", "key words": "hierarchical linear model, infrastructure performance, water treatment, system reliability, Senegal", "contribution statement": "