Classification of the existing knowledge base of OR/MS research and practice (1990–2019) using a proposed classification scheme

• Published in: Computers & operations research Vol. 118; pp. 104920 - 17
• Main Authors: Mustafee, Navonil, Katsaliaki, Korina
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01.06.2020; Pergamon Press Inc
• Subjects: Classification; Classification Scheme; Classification schemes; Decision analysis; Keywords; Knowledge bases (artificial intelligence); Operations research; OR/MS Education; OR/MS Knowledge Base; OR/MS Education; Classification Scheme; OR/MS Knowledge Base
• ISSN: 0305-0548; 1873-765X; 0305-0548
• DOI: 10.1016/j.cor.2020.104920

•We analyze keywords from over 82,000 articles published in 26 leading OR/MS journals (time frame of analysis is 30 years).•We describe a methodological approach for keyword selection and classification.•We present a hierarchal classification of 1300 most frequently used keywords and propose a classification scheme for OR/MS.•We present an analysis of OR/MS research and practice using the proposed classification scheme. Operations Research/Management Science (OR/MS) has traditionally been defined as the discipline that applies advanced analytical methods to help make better and more informed decisions. The purpose of this paper is to present an analysis of the existing knowledge base of OR/MS research and practice using a proposed keywords-based approach. A conceptual structure is necessary in order to place in context the findings of our keyword analysis. Towards this we first present a classification scheme that relies on keywords that appeared in articles published in important OR/MS journals from 1990–2019 (over 82,000 articles). Our classification scheme applies a methodological approach towards keyword selection and its systematic classification, wherein approximately 1300 most frequently used keywords (in terms of cumulative percentage, these keywords and their derivations account for more than 45% of the approx. 290,000 keyword occurrences used by the authors to represent the content of their articles) were selected and organised in a classification scheme with seven top-level categories and multiple levels of sub-categories. The scheme identified the most commonly used keywords relating to OR/MS problems, modeling techniques and applications. Next, we use this proposed scheme to present an analysis of the last 30 years, in three distinct time periods, to show the changes in OR/MS literature. The contribution of the paper is thus twofold, (a) the development of a proposed discipline-based classification of keywords (like the ACM Computer Classification System and the AMS Mathematics Subject Classification), and (b) an analysis of OR/MS research and practice using the proposed classification.