Panel-data econometric models are increasingly applied in engineering risk management to assess the effectiveness of safety interventions in industrial settings. Their methodological rigour in the specific context of manufacturing plants in developing economies requires critical evaluation. This study replicates and methodologically evaluates a panel-data framework for estimating risk reduction in manufacturing plants. The objective is to test the robustness of the estimation procedure, specifically examining model specification and the handling of unobserved heterogeneity. We replicate the analysis using an expanded dataset from Nigerian manufacturing plants. The core fixed-effects model is specified as $Y_{it} = \beta_0 + \beta_1 X_{it} + \alpha_i + \epsilon_{it}$, where $Y_{it}$ is the logged incident rate for plant $i$ in period $t$. Robust standard errors are clustered at the plant level to account for serial correlation. The replication confirms the negative association between capital investment in safety systems and reported incident rates, but the effect size is 40% smaller than originally estimated. Statistical inference shows the key coefficient remains significant at the 5% level, though with wider confidence intervals. The methodological evaluation underscores the sensitivity of risk reduction estimates to model specification and data quality. While the directional finding is robust, the magnitude of the effect is substantially attenuated under stricter econometric treatment. Future engineering risk studies employing panel data should prioritise diagnostic testing for fixed effects and heteroskedasticity. We recommend the collection of more granular, high-frequency operational data to improve model precision. replication study, risk management, panel data, fixed effects, manufacturing safety, econometric evaluation This study provides a novel, critical methodological appraisal of econometric techniques for engineering risk assessment, demonstrating that stricter replication protocols significantly alter the estimated efficacy of safety investments.