MACHINE OPTIMIZED NURSE ALLOCATION (MONA): TESTING THE FEASIBILITY OF AI-DRIVEN SCHEDULE AUTOMATION

ABSTRACTBACKGROUND: Manual nurse scheduling remains a persistent and underrecognized source of administrative burden for nurse leaders, particularly in high-acuity emergency department (ED) settings. Despite advances in healthcare technology, scheduling workflows often rely on manual processes that consume leadership time, contribute to burnout, and undermine workforce stability. This Doctor of Nursing Practice project examined the feasibility of using contemporary artificial intelligence (AI) tools to support or automate nurse scheduling through development and testing of a conceptual agent, Machine Optimized Nurse Allocation (MONA).METHODS: Guided by the Institute for Healthcare Improvement Model for Improvement, this feasibility project used iterative Plan–Do–Study–Act cycles to evaluate whether organization-approved AI platforms could ingest scheduling data, apply complex staffing rules, and reduce nurse leader administrative workload. The project was conducted in a freestanding emergency department within a large healthcare system. Baseline time tracking demonstrated that nurse leaders spent an average of 10 hours per week on scheduling tasks, most of which were interruptive and occurred outside scheduled work hours.RESULTS: Testing across multiple AI tools demonstrated that rule-based scheduling logic could be represented and applied in simulated environments. MONA accurately interpreted staffing policies, movement hierarchies, and scheduling rules when data was presented in structured text formats. However, system-level barriers prevented full automation, including lack of interoperability between scheduling software and AI platforms, irregular data formatting, restricted access to application programming interfaces, and enterprise security constraints that prevented live data ingestion and persistent agent memory.CONCLUSION: Although autonomous scheduling was not achievable within the current infrastructure, this project brought visibility to a largely invisible and unnecessary administrative burden carried by nurse leaders. Findings demonstrate this burden is not inherent to nursing leadership, but a consequence of outdated manual scheduling systems. By exposing its operational consequences, nurse scheduling was reframed as a strategic workforce issue rather than a clerical task. Most importantly, this feasibility work produced a practical implementation playbook documenting tested workflows, decision logic, system limitations, and prerequisites for future AI-driven scheduling. Addressing this burden through interoperable, workforce-centered technologies is essential to preserving nurse leader capacity and supporting sustainable emergency department operations.