"background": "The manufacturing sector is a critical component of Kenya's economic development strategy, yet systematic, data-driven methodologies for evaluating plant-wide operational efficiency remain underutilised. Current assessments often lack the statistical rigour to disentangle system-level from unit-level performance drivers. ", "purpose and objectives": "This short report aims to methodologically evaluate manufacturing plant systems and quantify efficiency gains using a multilevel modelling framework. The objective is to provide a replicable analytical approach for identifying significant levers for productivity improvement within the local industrial context. ", "methodology": "A cross-sectional study was conducted across multiple manufacturing plants. Operational data were collected on throughput, energy consumption, labour hours, and maintenance logs. A two-level hierarchical linear model was fitted, with production lines nested within plants. The core model is expressed as Y{ij = \0j + \1X1ij + eij, where \0j = \00 + \01Zj + u0j. Inference was based on robust standard errors. ", "findings": "The multilevel regression revealed that plant-level maintenance scheduling protocols accounted for approximately 22% of the variance in line-level efficiency. A significant positive association was found between predictive maintenance adoption and throughput (β = 0. 31, p < 0. 01). The intra-class correlation coefficient was 0. 18, indicating notable clustering effects. ", "conclusion": "The analysis confirms that a substantial portion of efficiency variation is attributable to plant-wide systemic factors, not just line-specific operations. The methodological approach successfully isolates these hierarchical influences. ", "recommendations": "Manufacturers should prioritise investments in centralised, data-informed maintenance systems. Further research should apply this modelling framework longitudinally to assess causal impacts of specific interventions. ", "key words": "industrial engineering, operational efficiency, hierarchical linear model, predictive maintenance, sub-Saharan Africa", "contribution statement": "This report provides a
Mwangi et al. (Fri,) studied this question.