Enhancing Decision Support in Business Information Systems through The Integration of Generative AI Models | ||
| Journal of Modern Business and Technology | ||
| Articles in Press, Accepted Manuscript, Available Online from 01 November 2025 PDF (724.51 K) | ||
| Document Type: Research Article | ||
| DOI: 10.21608/jmbt.2025.432445.1002 | ||
| Authors | ||
| Mariam K. Bersi* 1; Ahmed M. Eassa2 | ||
| 1Teaching Assistant, Business Informatics Department, Faculty of Management and Business Intelligence, MTI University | ||
| 2Assistant Professor, Business Informatics Department, Faculty of Management and Business Intelligence, MTI University | ||
| Abstract | ||
| The rapid rise of Generative AI in enterprise applications has made its responsible integration into Business Information Systems (BIS) a critical priority. This study proposes a governance-aware, five-layer architecture that integrates Retrieval-Augmented Generation (RAG), explainable AI (XAI), and human-in-the-loop (HITL) safeguards to address the shortcomings of traditional decision-support systems. The architecture was implemented using a hybrid technology stack and evaluated with 45 participants across multiple business domains. Results showed a 93% reduction in hallucinations, a 58% improvement in task completion time, and almost double the user acceptance rate. These outcomes highlight the framework’s ability to deliver measurable business value while maintaining transparency, trust, and compliance, positioning it as a practical solution for enterprise-wide adoption. Beyond performance gains, the study emphasizes the importance of governance and explainability in building user confidence. Overall, the framework contributes a replicable and enterprise-ready model that balances innovation with accountability in BIS contexts. | ||
| Keywords | ||
| Business Information Systems; Retrieval-Augmented Generation (RAG); Explainable AI (XAI); Human-in-the-Loop (HITL); Governance | ||
|
Statistics Article View: 1,227 PDF Download: 473 |
||
