A Hybrid Business Intelligence–Machine Learning Model for Proactive Cost Reduction in Healthcare Operations
Year: 2025
Type: article
Abstract: Healthcare organisations are being increasingly pressured to cut the costs of operations without compromising on service quality. Conventional Business Intelligence (BI) systems have been effective in organizing data to monitor performance and enhance planning; however, they are limited to retrospective analysis. Conversely, Machine Learning (ML) has proven to be highly predictive, enabling hospitals to forecast admissions, treatment expenses, and inefficiencies. Nevertheless, these two techniques are commonly used separately, thereby restricting their potential for transforming healthcare decision-making. The present work develops and tests a hybrid BIML framework that enables the incorporation of descriptive, diagnostic, and predictive analytics into a single decision-support system. Based on operational cost data, hospital electronic health records, and claims, the study utilizes BI dashboards for performance monitoring and applies ML algorithms to forecast costs and identify inefficiencies. The hybrid model is trialed in finance, operations, and clinical departments to gather varied knowledge on resource utilization. Three criteria are used to assess the effectiveness of the model, determining its predictive accuracy, cost savings, and efficiency gains. The hybrid system is compared with BI-only and ML-only systems to demonstrate the value addition of the hybrid system. The results of this research are expected to indicate that the combination of BI and ML will enable the transformation of healthcare decision-making from reactive to proactive, resulting in quantifiable cost savings and improved operational outcomes.
Author: Eunice Feyisayo Ogundipe
Language: English
Subfield: Management Information Systems
Domain: Social Sciences
Open Access status: bronze
OpenAlex ID: W4415731311
FWCI:
Cites:
Cited by:
Related to: