"background": "Manufacturing systems in developing economies face unique operational and infrastructural challenges that heighten systemic risk. Current risk assessment methodologies often lack the granularity to account for the hierarchical structure of plant-level data, where observations within a single facility are correlated. ", "purpose and objectives": "This case study aims to methodologically evaluate risk factors within Ghanaian manufacturing plants and quantify the efficacy of targeted interventions in reducing systemic operational risk. The primary objective is to develop and apply a multilevel modelling framework suitable for clustered industrial data. ", "methodology": "A case study was conducted across multiple manufacturing plants. A two-level random intercepts regression model was fitted to assess risk reduction. The core statistical model is expressed as Risk{ij = \0 + \1Xij + uj + eij, where i denotes observations and j denotes plants. Robust standard errors were used for inference to account for heteroscedasticity. ", "findings": "The multilevel analysis revealed that procedural standardisation interventions yielded a significant reduction in composite risk scores. Specifically, plants implementing the full intervention protocol showed a mean risk reduction of 22. 4% (95% CI: 18. 1% to 26. 7%) compared to control groups. The intra-class correlation coefficient indicated that 31% of the variance in risk was attributable to differences between plants. ", "conclusion": "The applied multilevel regression provides a superior methodological framework for evaluating nested manufacturing data, confirming that structured procedural interventions are effective for systemic risk mitigation in this context. ", "recommendations": "Manufacturing plant managers should adopt hierarchical data analysis to inform decision-making. Policymakers are advised to support the development of context-specific operational standards that facilitate procedural standardisation across the sector. ", "key words": "multilevel modelling, operational risk, manufacturing systems, regression analysis, industrial engineering, Ghana", "contribution statement": "This study provides a novel application of multilevel regression to manufacturing
Kwame Asante (Wed,) studied this question.