Abstract
{ "background": "The systematic evaluation of manufacturing plant systems in sub-Saharan Africa is limited, with a paucity of rigorous field data on operational efficiency and cost-effectiveness. This gap hinders evidence-based investment and policy decisions in the region's industrial sector.", "purpose and objectives": "This study aimed to methodologically evaluate and compare the cost-effectiveness of two distinct manufacturing plant systems—a modular, semi-automated system versus a conventional, labour-intensive setup—within the Ugandan context.", "methodology": "A randomised field trial was conducted across multiple manufacturing sites. Plants were randomly assigned to implement one of the two systems. Cost and output data were collected over an operational period. Cost-effectiveness was analysed using a generalised linear model: $\\ln(\\text{Cost}{it}) = \\beta0 + \\beta1 \\text{System}i + \\beta2 \\text{Output}{it} + \\epsilon_{it}$, with robust standard errors clustered at the plant level.", "findings": "The modular system demonstrated superior cost-effectiveness, reducing unit production costs by an average of 18.7% (95% CI: 12.3% to 25.1%) compared to the conventional system, after controlling for scale. This difference was statistically significant (p < 0.01).", "conclusion": "The modular, semi-automated plant system presents a more cost-effective solution for manufacturing in the studied setting, primarily through significant reductions in variable costs per unit.", "recommendations": "Manufacturing investors and policymakers should prioritise support for and adoption of modular, semi-automated systems. Further research should investigate the long-term maintenance costs and scalability of such systems.", "key words": "manufacturing systems, cost-effectiveness, randomised field trial, industrial engineering, sub-Saharan Africa", "contribution statement": "This paper provides the first application of a randomised field trial methodology to compare engineering plant systems in an African industrial context, generating novel, high-certain