Scholarlink

<p>Agentic artificial intelligence (AI), characterized by autonomous, goal-oriented systems capable of inter-agent collaboration, is rapidly gaining traction as a transformative paradigm in the context of small, medium, and micro enterprises (SMMEs). The conceptual foundations and applied models of agentic AI have been thematically reviewed, but the structural and intellectual evolution of this domain needs further exploration. This bibliometric review builds on a previously conducted systematic literature review (SLR) of 66 studies and expands the analysis with an updated dataset of 214 papers published between 2019 and 2025, sourced from Scopus, Google Scholar, and credible gray literature repositories. The study uses VOSviewer and the Bibliometrix R package to create scholarly trend maps, identify prominent contributors, thematic clusters, and existing gaps in the literature. Findings reveal a sharp post-2022 surge, with 162 publications from 2023 to mid-2025, driven by developments in large language models (LLMs) and multi-agent systems (MAS). Key themes include foundational AI techniques such as LLMs and reinforcement learning (RL), strategic applications of agentic AI in SMMEs, decision-support systems, simulation-based modeling, and industrial implementations including Industry 4.0 and cyber-physical systems. Geographically, the majority of agentic AI research originates in the United States, China, and the United Kingdom, indicating a concentrated intellectual influence in affluent nations. In contrast, contributions from regions such as Africa and Southeast Asia remain limited. Substantial gaps persist, including limited empirical implementations in micro enterprises, underdeveloped AI governance frameworks, a lack of region and sector-specific adaptations, insufficient analysis of adoption and, fragmented interdisciplinary collaboration, guiding future research and policy in resource-constrained SMME contexts.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Mapping the research landscape of agentic AI in SMMEs through a bibliometric analysis of patterns and knowledge gaps

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月25日

查看全文
  • Peter Adebowale Olujimi,
  • Pius Adewale Owolawi,
  • Agnieta Pretorius,
  • Etienne Van Wyk

摘要

Agentic artificial intelligence (AI), characterized by autonomous, goal-oriented systems capable of inter-agent collaboration, is rapidly gaining traction as a transformative paradigm in the context of small, medium, and micro enterprises (SMMEs). The conceptual foundations and applied models of agentic AI have been thematically reviewed, but the structural and intellectual evolution of this domain needs further exploration. This bibliometric review builds on a previously conducted systematic literature review (SLR) of 66 studies and expands the analysis with an updated dataset of 214 papers published between 2019 and 2025, sourced from Scopus, Google Scholar, and credible gray literature repositories. The study uses VOSviewer and the Bibliometrix R package to create scholarly trend maps, identify prominent contributors, thematic clusters, and existing gaps in the literature. Findings reveal a sharp post-2022 surge, with 162 publications from 2023 to mid-2025, driven by developments in large language models (LLMs) and multi-agent systems (MAS). Key themes include foundational AI techniques such as LLMs and reinforcement learning (RL), strategic applications of agentic AI in SMMEs, decision-support systems, simulation-based modeling, and industrial implementations including Industry 4.0 and cyber-physical systems. Geographically, the majority of agentic AI research originates in the United States, China, and the United Kingdom, indicating a concentrated intellectual influence in affluent nations. In contrast, contributions from regions such as Africa and Southeast Asia remain limited. Substantial gaps persist, including limited empirical implementations in micro enterprises, underdeveloped AI governance frameworks, a lack of region and sector-specific adaptations, insufficient analysis of adoption and, fragmented interdisciplinary collaboration, guiding future research and policy in resource-constrained SMME contexts.