Deployment of Models into Adaptive Business Intelligence Systems for Hospital Intelligent Scheduling

The scheduling of medical examinations involves coordinating multiple operational constraints, including limited capacity, work shifts, and the need to accommodate manual interventions in the planning process. In this context, predictive models and optimization methods have been explored as planning support tools, but they require appropriate mechanisms for integration into decision-support systems. This article describes the development and implementation of a decision-support system for intelligent examination scheduling in a hospital setting, framed within an Adaptive Business Intelligence (ABI) approach. The solution integrates previously developed analytical components, namely predictive models and optimization methods, organized within a modular architecture and exposed through an API. The system uses operational data to generate scheduling proposals compatible with real-world constraints, such as shift definitions, simultaneous examination capacity, and the presence of manually scheduled procedures. These proposals are produced by a configurable optimization module and stored as structured objects, enabling scenario comparison and iterative validation of solutions.