Home
This study evaluated whether a reproducible Python workflow can strengthen Data Envelopment Analysis in industrial efficiency studies when compared with spreadsheet-based tools such as XlDEA/XIDEA. A methodological, documentary, and computational comparative design was applied. The study examined two implementation environments: a reproducible Python workflow and spreadsheet-based analysis tools. Data were collected through a structured comparison matrix that assessed methodological coverage, automation and scalability, reproducibility and auditability, and second-stage inferential robustness. The analytical procedure reviewed input-oriented CCR estimation, bootstrap inference, Tobit modeling on inefficiency, truncated regression with double bootstrap, and automated report generation. The main result indicates that Python provides a more scalable and auditable architecture for repeated analysis, especially when monthly data, multiple decision-making units, and standardized outputs are required. However, spreadsheet tools remain useful for exploratory applications because they offer greater initial accessibility for non-programming users. The study concludes that Python is preferable for production-grade efficiency analysis, while truncated regression with double bootstrap should guide future second-stage inference when contextual determinants of efficiency are analyzed.