Artificial Intelligence for Hospital Scheduling: A Simple, Reproducible, and Effective Model

TL;DR

This paper introduces an AI-based hospital scheduling system using Excel and ChatGPT to reduce administrative work and improve fairness.

Contribution

A novel, accessible AI-driven hospital scheduling model using natural language and Excel automation.

Findings

The AI system reduced administrative time and improved schedule fairness.

The model minimized team conflicts and enhanced staff satisfaction.

The system is simple, reproducible, and adaptable to other sectors.

Abstract

Hospital scheduling, particularly for on-call shifts and daily assignments, is a complex task that must account for numerous factors, such as service requirements, staff preferences, and unplanned absences. Traditional methods often result in significant administrative burden and can lead to staff frustration, potentially affecting the quality of care. This study explores the use of artificial intelligence-based large language models (AI LLMs) to automate hospital scheduling through widely accessible tools, aiming to simplify the process, reduce manual effort, and enhance fairness.

ChatGPT® (OpenAI, San Francisco, CA, USA) is used to translate natural language instructions into VBA (Visual Basic for Applications) macros, which automate the creation of on-call and daily activity schedules. The process involves collecting staff preferences via a Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) sheet, followed by AI-generated VBA macros that automate the creation of the schedule, ensuring adherence to various constraints such as equitable shift distribution and prioritization of specific roles. The system was developed by a non-IT professional and does not require advanced programming skills.

The implementation of this AI-driven scheduling system resulted in a significant reduction in administrative time and increased schedule fairness, as decisions were based on clear, consistently applied rules. The system also minimized conflicts within teams, improving both organizational efficiency and staff satisfaction. However, the development of such a process was not without its challenges, particularly in terms of rule formulation and Excel cell references.

The integration of Microsoft Excel® and AI LLM provides a simple and reproducible solution for hospital schedule organization, reducing administrative burden and promoting fairness. This model, which can be adapted to other sectors facing similar challenges, enables teams to retain control over the process.

Figures