Abstract
{ "background": "The adoption of advanced manufacturing systems is critical for industrial development, yet robust methodologies for tracking and forecasting their uptake in emerging economies are lacking. Rwanda's strategic industrial policy provides a pertinent case for methodological development in this domain.", "purpose and objectives": "This study aims to methodologically evaluate frameworks for assessing manufacturing systems and to estimate adoption rates using a panel-data model. The objective is to provide a replicable analytical tool for engineering and policy analysis.", "methodology": "A longitudinal dataset from a census of medium and large plants was constructed. Adoption was modelled using a fixed-effects panel regression: $A{it} = \\alphai + \\beta1Tt + \\beta2X{it} + \\epsilon{it}$, where $A{it}$ is the adoption status for plant $i$ at time $t$, $\\alphai$ denotes plant-specific effects, $T$ is a time trend, and $X$ is a vector of plant-level covariates. Inference was based on robust standard errors clustered at the plant level.", "findings": "The methodological evaluation identified significant measurement biases in conventional survey tools. The panel estimation revealed a positive, statistically significant time trend ($\\beta1 = 0.07$, 95% CI [0.04, 0.10]), indicating an average annual increase in adoption likelihood of 7 percentage points. Plant size and export orientation were key determinants.", "conclusion": "The proposed panel-data methodology offers a more reliable alternative for measuring technological diffusion. The results confirm a steady, albeit moderate, uptake of modern manufacturing systems, driven by specific firm characteristics.", "recommendations": "Industrial policy should target support towards smaller, non-exporting firms to broaden adoption. Future research should integrate the developed methodology with techno-economic feasibility assessments.", "key words": "manufacturing systems, technology adoption, panel data, fixed effects, industrial policy, econometrics", "contribution statement": "This paper provides a novel panel-data estimation