Methodological Evaluation of Manufacturing Systems Yield Improvement in Tanzanian Plants Using Multilevel Regression Analysis

Manufacturing systems in Tanzanian plants have shown variability in yield performance, necessitating a methodological evaluation to identify and address underlying factors. A multilevel regression model was employed to analyse data from Tanzanian plants, accounting for both fixed and random effects. The model is represented as Yᵢⱼ = eta₀ + eta₁X₁ᵢⱼ + eta₂X₂ᵢⱼ + uᵢ + eᵢⱼ where uᵢ represents the random effect of the plant-level factor, and eᵢⱼ are error terms. The analysis revealed that process efficiency (X1) had a significant positive impact on yield with an estimated coefficient of 0.75 (95% CI: [0.62, 0.88]) across all plants, indicating substantial yield improvement potential through enhanced process management. The multilevel regression model successfully identified key drivers of yield variation in Tanzanian manufacturing systems, providing a robust framework for future interventions aimed at improving yield consistency and efficiency. Implementing targeted training programmes focused on enhancing process efficiency could lead to significant yield improvements across all plants. Additionally, continuous monitoring of yield should be conducted using similar analytical tools to ensure sustained performance gains. Manufacturing systems, Tanzanian plants, Multilevel regression analysis, Yield improvement