Supplementary data and code for: A Decade of Metaheuristic Optimization in Process Systems Engineering (2015–2025): A Critical Systematic Review

This repository contains the data, protocols, and analysis code underlying the systematic review "A Decade of Metaheuristic Optimization in Process Systems Engineering (2015–2025): A Critical Systematic Review" by Fernando Israel Gómez-Castro and Oscar Daniel Lara-Montaño. The study is a PRISMA 2020 systematic review of metaheuristic optimization in process systems engineering (PSE). A search of Web of Science and Scopus (January 2015 to December 2025, executed in April 2026) produced a final corpus of 808 articles published in 58 peer-reviewed journals. The entire corpus was classified by primary algorithm, algorithm family, PSE application domain, objective formulation, hybrid type, and algorithmic novelty. A stratified subsample of 103 articles underwent full-text analysis of methodological practices — benchmarking, statistical testing, reproducibility, and parameter reporting — and was scored with the proposed Metaheuristic Reporting Quality Index (MRQI), an eight-indicator scale from 0 to 10. The materials provided here allow the corpus statistics, the methodological-quality assessment, the MRQI, and the temporal trend analyses reported in the article to be reproduced. Contents:- data/ — the 808-article corpus table with classification metadata; the full-text extraction dataset for the 103-article subsample; the per-article MRQI scores; the list of 58 retained journals; the Cochran–Armitage trend-test results; and the raw inter-rater reliability scores.- protocols/ — the completed PRISMA 2020 checklist; the complete Boolean search queries for Web of Science and Scopus; the full-text data-extraction protocol; and the MRQI scoring rubric.- scripts/ — Python scripts for the algorithm and family reclassification, the corpus totals, the Cochran–Armitage trend tests, the MRQI computation, the inter-rater reliability statistics, and the figure generation. A README file describes every file in detail. Not included: the full texts (PDFs) of the 808 reviewed articles are not redistributed, as they remain under their publishers' copyright; each article is identified by its DOI for retrieval from the original source. Raw bibliographic database exports are also omitted because they contain publisher-supplied abstracts; the exact search queries needed to regenerate them are provided in this repository. When reusing these materials, please cite both the associated journal article and this repository.