Capability Maturity Models Across Industries: A Global Bibliometric Analysis of Intellectual Evolution, Thematic Clusters, and Emerging Frontiers (1992–2025)
Description
The Study contains a bibliometric analysis of the global research on Capability Maturity Model (CMM) (1992-2025) conducted to map the intellectual development, topical organisation and research frontiers of the field across industries. A systematic bibliometric analysis of 500 journal articles indexed by Scopus, extracted from a sample of 4,474 records, based on specific inclusion criteria (article type, keyword match, English language). Keywords co-occurrence mapping and visualisation, as well as cluster analysis, were performed using VOSviewer, and Kleinberg’s burst detection was used to identify research bursts. Silhouette scores were used to assess cluster validity and thematic coherence. Other performance indicators, including citations, h-index, and citation density were calculated to assess intellectual performance, while geographic and source mapping assessed the worldwide spread of research. The dataset covers 13,675 citations, an h-index of 58, and an average citation density of 27.35, indicating a mature and impactful field. The US is the main source, with Norway, Portugal and Turkey showing strong citation density through digital transformation research. Keyword burst activity shows four major areas of focus; digital transformation, Industry 4.0, cybersecurity and cross-domain maturity modelling - with significant activity from 2020 to 2025, indicating a transition to digital capability models. This is a cross-industry bibliometric analysis of CMM research, showing its evolution into a generalised digital capability assessment
Files
Steps to reproduce
The Scopus database was selected for its comprehensive and consistent coverage of peer-reviewed journals in the fields of engineering, computer science, medicine and health sciences from the early 1990s onwards, which is a time span indispensable for the study of 33 years of intellectual development. The search string was: TITLE-ABS-KEY ( Capability Maturity Models ) AND PUBYEAR > 1989 AND PUBYEAR < 2026 AND ( LIMIT-TO ( SUBJAREA , "ENGI" ) OR LIMIT-TO ( SUBJAREA , "MEDI" ) OR LIMIT-TO ( SUBJAREA , "HEAL" ) OR LIMIT-TO ( SUBJAREA , "PHAR" ) OR LIMIT-TO ( SUBJAREA , "NURS" ) OR LIMIT-TO ( SUBJAREA , "NEUR" ) OR LIMIT-TO ( SUBJAREA , "DENT" ) OR LIMIT-TO ( SUBJAREA , "VETE" ) OR LIMIT-TO ( SUBJAREA , "ENVI" ) OR LIMIT-TO ( SUBJAREA , "COMP" ) ) AND ( LIMIT-TO ( DOCTYPE , "ar" ) ) AND ( LIMIT-TO ( EXACTKEYWORD , "Maturity Model" ) OR LIMIT-TO ( EXACTKEYWORD , "Capability Maturity Models" ) OR LIMIT-TO ( EXACTKEYWORD , "Capability Maturity Model" ) OR LIMIT-TO ( EXACTKEYWORD , "Capability Maturity Model Integration" ) OR LIMIT-TO ( EXACTKEYWORD , "Capability Maturity Model (cmm)" ) OR LIMIT-TO ( EXACTKEYWORD , "Maturity Levels" ) ) AND ( LIMIT-TO ( LANGUAGE , "English" ) ) The initial retrieval yielded 4,474 documents. Sequential filtering was applied: only documents in engineering, computer science, environmental science, medicine, health profession, pharmacology, nursing, neuroscience, dentistry and veterinary subject areas (3,519 documents); only journal articles (1,393 documents); keyword filtering to ensure documents explicitly discuss CMM concepts (529 documents); only documents published in English (500 documents for further analysis). Six analytical methods were used on the 500 articles. First, trend analysis identified the development of the field over time according to the pattern of annual publications over 33 years in four phases. Second, geographic analysis characterised the distribution of publications and citations by country of author affiliation. Third, author productivity analysis characterised the most prolific authors by the number of documents and citations they received. Fourth, publication productivity and citation impact analysis identified journals publishing knowledge in the field, including citations per paper. Fifth, we analysed the co-occurrence network of keywords in VOSviewer using a minimum occurrence of 10, generating a cluster network from the 500-document sample after calibrating the parameters as per Aghimien et al. (2020) and Klarin (2024). Sixth, keyword burst analysis used Kleinberg's (2002) two-state automaton approach to all keywords that occur at least three times within the 33-year observation window, calculating burst strength and silhouette scores.
Institutions
- Durban University of TechnologyKwaZulu-Natal, Durban