"background": "Time-series forecasting models for cost-effectiveness diagnostics are critical for operational efficiency in industrial settings. A previously proposed model for manufacturing systems has been cited in policy discussions, yet its empirical robustness and generalisability within the specific context of domestic plants require rigorous verification. ", "purpose and objectives": "This study aimed to independently replicate and validate the specified autoregressive integrated moving average (ARIMA) forecasting model using original and extended operational data from a representative sample of plants. The objective was to assess its predictive accuracy and practical utility for diagnostic cost analysis. ", "methodology": "We executed a direct computational replication followed by a measurement replication using a newly collated dataset of plant-level operational metrics. The core model, specified as Ct = \ + =1^{p\ Ct-i + \ + =1^q\ -j, where Ct is the cost-effectiveness index, was re-estimated. Predictive performance was evaluated using mean absolute percentage error (MAPE) and Diebold-Mariano tests against benchmark models. ", "findings": "The replication confirmed the model's structural form but revealed a consistent upward bias in its medium-term forecasts, with predicted cost-efficiency gains approximately 15% higher than observed values. The 95% confidence interval for the key autoregressive parameter \1 did not contain the original point estimate, indicating a statistically significant difference in the underlying process. ", "conclusion": "While the model's theoretical framework remains sound, its calibrated parameters are not directly transferable, highlighting the sensitivity of such models to localised operational conditions and data structures. The original findings are partially supported but require contextual recalibration. ", "recommendations": "Future applications must involve recalibration with local data prior to deployment. We recommend the model be integrated with exogenous variable analysis to improve its stability and provide a protocol for periodic parameter updating for plant managers. ", "key words": "replication study,
Suleiman et al. (Tue,) studied this question.