Artificial intelligence and end user tools to develop a nurse duty roster scheduling system.

Area of Science:

• Healthcare Management
• Operations Research
• Artificial Intelligence

Background:

• Manual nurse duty roster preparation is time-consuming, often exceeding a full workday for ward managers.
• Nurse shift requests and hospital policies complicate manual scheduling, impacting efficiency and staff satisfaction.
• Existing scheduling methods may lack flexibility and optimal resource allocation.

Purpose of the Study:

• To develop an automated nurse roster scheduling system.
• To integrate open-source operational research tools for efficient scheduling.
• To consider hospital standards and nurse-specific constraints in the scheduling process.

Main Methods:

• Utilized artificial intelligence and operational research tools to develop the scheduling system code.
• Employed an open architecture design for adaptability and comparison with existing heuristics.
• Integrated real-world shift duty requirements from a hospital ward setting.

Main Results:

• Schedules generated within one minute, optimizing both hard and soft constraints.
• Achieved full compliance with all hard constraints and high satisfaction of nurse soft constraints.
• Computer-generated schedules demonstrated superior optimization and flexibility compared to manual methods, ensuring safe staffing levels.

Conclusions:

• An economical, efficient, and user-friendly nurse scheduling system was developed using open-source operational research.
• The open-source platform proved effective for scheduling applications and can be implemented across hospital wards.
• The system is adaptable for updates with new policies and manpower changes by trained hospital IT personnel.