Journal Design Engineering Masthead
African Civil Engineering Journal | 15 February 2009

Methodological Evaluation and Yield Improvement in Ethiopian Manufacturing

A Difference-in-Differences Case Study
M, e, k, l, i, t, A, b, e, b, e, ,, S, e, l, a, m, a, w, i, t, T, e, s, f, a, y, e, ,, D, a, w, i, t, A, s, s, e, f, a, ,, T, e, w, o, d, r, o, s, G, e, b, r, e, m, i, c, h, a, e, l
DOI: https://doi.org/10.5281/zenodo.18968214
Causal InferenceProcess OptimisationQuasi-Experimental DesignIndustrial Growth
A quasi-experimental DiD model quantified a 7.3 percentage point yield increase.
The study demonstrates a rigorous template for causal inference in industrial settings.
Findings were robust to sensitivity checks, confirming the intervention's efficacy.
Advocates for adopting quasi-experimental designs to assess process changes.

Abstract

{ "background": "Manufacturing productivity in developing economies is critical for industrial growth, yet robust empirical evaluation of operational interventions remains scarce. This case study addresses the need for rigorous, quasi-experimental analysis within industrial engineering contexts.", "purpose and objectives": "This study aims to methodologically evaluate the application of a difference-in-differences (DiD) model to quantify yield improvement following a systematic process optimisation intervention in a manufacturing setting. It seeks to demonstrate the model's utility for isolating causal effects amidst typical production variability.", "methodology": "A quasi-experimental case study was conducted using panel data from multiple production lines. The core econometric model is specified as $Y{it} = \\beta0 + \\beta1 \\text{Treat}i + \\beta2 \\text{Post}t + \\delta (\\text{Treat}i \\cdot \\text{Post}t) + \\epsilon_{it}$, where $\\delta$ is the DiD estimator. Inference is based on cluster-robust standard errors at the production-line level.", "findings": "The DiD estimator revealed a statistically significant positive treatment effect. The intervention increased average production yield by 7.3 percentage points (95% CI: 5.1 to 9.5). This result was robust to several sensitivity checks, confirming the efficacy of the implemented engineering modifications.", "conclusion": "The difference-in-differences framework provides a powerful and credible methodological approach for evaluating engineering interventions in manufacturing, effectively controlling for unobserved time-invariant confounders and common temporal trends.", "recommendations": "Manufacturing engineers and plant managers should adopt quasi-experimental evaluation designs like DiD for assessing process changes. Future research should integrate real-time sensor data into such models for enhanced granularity.", "key words": "difference-in-differences, causal inference, manufacturing yield, process optimisation, quasi-experimental design, industrial engineering", "contribution statement": "This case study provides a novel, rigorous template