There are a number of limitations that should be mentioned explicitly․ First‚ in this analysis‚ the study relied on a specific database and defined time window‚ work published outside this window was located in non-indexed/journals in the respective countries languages other than English‚ or as working papers and preprints may not have been captured․ The clustering tools used (Biblioshiny) to group papers according to their citation and co-occurrence networks․ The thematic interpretations here are based on data-driven approximations rather than expert judgment․than definitive‚ expert-validated categorizations․ Since this field has only been in its early-to-mid growth phase‚ expert-validated categorizations of the life-cycle model predicting a peak around 2033‚ and patterns relative to emerging AI-driven natural language themes‚ are likely to shift as more publications emerge over the next few years․ The approach to analysis is quantitative and pattern-based‚ showing how much is published‚ where‚ and by which researchers‚ but does not evaluate the methodological quality or real-world predictive effectiveness of the OR techniques studied‚ so a The growing number of publications on the theme could be due to research interest․Finally‚ as is the case with any bibliometric analysis‚ the quality of the study is limited by the quality of the underlying metadata: author names‚ affiliations‚ and keywords tagging are all subject to some inconsistencies that introduce slight distortions into such as Lotka's law and Bradford's Law․