The Evolving Architecture of Insight: A Holistic Survey of Next-Generation Business Intelligence and Analytics

Abstract

Business Intelligence and Analytics has transformed from static historical reporting into a dynamic, real-time discipline at the intersection of data management, machine learning, and decision science. This paper presents a comprehensive survey of the evolving BIA landscape, examining both back-end architectural innovations and front-end analytical capabilities. We trace the progression from traditional three-tier BI systems to next-generation paradigms including operational BI for real-time insights, situational BI for integrating external streaming data, and self-service BI for democratized analytics. The survey reviews enabling database technologies such as in-memory databases and hybrid OLTP/OLAP systems, and addresses critical data governance challenges arising from heterogeneous data sources and increased user participation. On the analytics front, we examine machine learning applications for time series forecasting across financial, sales, and healthcare domains, and highlight integrative approaches, particularly multiple kernel learning for combining disparate data sources to improve predictive accuracy. A healthcare case study in intensive care unit monitoring demonstrates how next-generation BIA can integrate real-time sensor data with traditional records for life-critical decision support. This survey provides researchers and practitioners with a holistic framework for understanding the converging technologies shaping the future of data-driven enterprise intelligence.