From Industrial Engineering to Elderly Care: A Thematic Review of OR and AI Integration

As the global population ages rapidly, healthcare systems face severe challenges in efficiency, equity, and patient-centered care. Industrial engineering and operations research offer powerful optimization and decision-support tools, but current applications remain fragmented. This review selects 30 representative studies categorized into three themes: home healthcare operations (e.g., nurse scheduling and routing), polypharmacy management (medication adherence optimization), and clinical chronotherapy (personalized treatment timing). Methodologically, early studies used static linear programming, while recent work shifts to dynamic AI methods like reinforcement learning and stochastic optimization. However, the vast majority focus on operational metrics (e.g., cost, time) and rarely evaluate real clinical outcomes like discharge rates, satisfaction, or adverse events. Moreover, integrated models for hospital-to-community transitions are critically lacking. To address this, the paper proposes a conceptual framework combining digital twins, LLM reasoning, and human feedback for adaptive decisions from task to system level. The framework emphasizes robustness analysis, human-centric design, and technology empowerment to translate theoretical metrics into tangible health improvements.