Citation

Shoaib, Muhammad and Mohmand, Muhammad Ismail and Sayed, Nasir and Khan, Inayat and Jan, Salman and Lee, It Ee (2026) AI-Powered Human Resource Management System for Hospitals: Enhancing Workforce Planning, Scheduling, and Performance Evaluation. IEEE Access. p. 1. ISSN 2169-3536

Text 3.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Efficient and adaptive workforce management is a critical component of hospital operations, especially in high-demand and dynamic healthcare environments. This research presents the design, development, and evaluation of an AI-powered Human Resource Management System tailored for hospitals. The system integrates multiple artificial intelligence components—including time-series forecasting, heuristic optimization, explainable machine learning, and real-time visualization—to automate and enhance core human resource functions such as staffing prediction, shift scheduling, performance evaluation, and decision support. The forecasting module leverages Long Short-Term Memory networks to predict departmental staffing needs up to seven days in advance using historical inflow, shift patterns, and contextual factors such as public holidays. A genetic algorithm based scheduling engine generates weekly shift plans optimized for staff satisfaction, regulatory compliance, and operational efficiency. Staff performance is assessed using a Random Forest classifier supported by SHapley Additive Explanations to ensure interpretability and transparency in tier-based classification. Furthermore, a context-aware decision support layer enables real-time adaptation to anomalies such as absenteeism or public health emergencies using rule-based logic and anomaly detection. The system was implemented as a cloud-deployed dashboard using Flask and ReactJS, providing human resource managers with interactive tools for schedule editing, forecast review, and performance insight. Experimental evaluations demonstrate high accuracy in forecasting, reduced overtime costs, improved staff satisfaction, and scalability for concurrent users. Experimental results show that the proposed system achieves up to 23% reduction in overtime hours, improves staff satisfaction by approximately 5%, attains forecasting accuracy with an average R² score of 0.895 over a 7-day horizon, and supports up to 50 concurrent users with acceptable latency. The proposed AI-HRMS sets a foundation for intelligent, responsive, and explainable workforce management in healthcare, offering substantial improvements over traditional HR practices. Future work will focus on live deployment, reinforcement learning integration, and federated collaboration across institutions.

Item Type:	Article
Uncontrolled Keywords:	AI-HRMS, Workforce Forecasting, Smart Scheduling, Explainable AI, Hospital Resource Management
Subjects:	R Medicine > R Medicine (General) > R855-855.5 Medical technology
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	04 Jun 2026 05:57
Last Modified:	04 Jun 2026 05:57
URII:	http://shdl.mmu.edu.my/id/eprint/15931

Downloads

Edit (login required)