Abstract
Business Intelligence and Analytics (BI&A) systems are essential for data-driven decision making. Their application in developing economies remains limited due to infrastructure gaps, scarce expertise and institutional fragmentation. Recent progress in large language models (LLMs) offers new ways to extend BI&A systems through natural language processing. This paper proposes a theoretical framework for LLM-enhanced BI&A systems to support environmental monitoring in developing economies. The framework draws on Information and Communication Technologies for Development (ICT4D), Information Foraging Theory (IFT) and Socio-Technical Systems Theory (STST). It emphasizes the need to align technical design with institutional capacity. Preliminary findings of our case study of the Nepalese cement industry demonstrate applicability of the framework and LLM-enhanced BI&A systems to addressing the multi-layered needs for environmental monitoring.
Recommended Citation
Karn, Sarthak and Chung, Wingyan, "Designing LLM-Enhanced Business Intelligence and Analytics Systems: The Case of the Nepalese Cement Industry" (2025). GlobDev 2025. 6.
https://aisel.aisnet.org/globdev2025/6