{ "background": "Community health centres in Ethiopia face significant operational risks from supply chain disruptions and variable patient demand, which compromise service delivery and public health outcomes. Existing planning tools often lack the predictive rigour needed for proactive resource allocation.", "purpose and objectives": "This case study aims to methodologically evaluate the application of a novel time-series forecasting model to measure and reduce operational risk in these centres. The objective is to assess the model's predictive performance and its utility for strategic health resource planning.", "methodology": "We developed and applied a seasonal autoregressive integrated moving average (SARIMA) model, formalised as $\\phi(B)\\Phi(B^s)\\nabla^d\\nabla^Ds Yt = \\theta(B)\\Theta(B^s)\\epsilon_t$, to historical centre-level data on essential medicine stock-outs and outpatient attendance. Model diagnostics included analysis of robust standard errors and Ljung-Box tests for residual autocorrelation.", "findings": "The model demonstrated strong predictive accuracy for monthly medicine stock-out rates, with a mean absolute percentage error of 8.7%. Forecasts indicated a persistent seasonal pattern, with a 22% increase in stock-out risk projected for the late-year rainy season period. Uncertainty intervals widened notably beyond an 18-month forecast horizon.", "conclusion": "The SARIMA modelling approach provides a statistically sound methodology for quantifying and anticipating operational risks in low-resource community health settings. It offers a substantial improvement over reactive management practices.", "recommendations": "Health centre managers should integrate such forecasting into quarterly logistical planning. National programmes should support the development of centralised forecasting units and build analytical capacity among district health officers.", "key words": "forecasting, health systems, operational risk, SARIMA, supply chain, public health planning", "contribution statement": "This study provides a novel, transferable methodological framework for applying time-series analytics to measure health system risk, demonstrating its utility through a concrete application that reduced forecast error for stock-outs by over 30% compared to