Optimization Techniques for Physician Scheduling Problem: A Systematic Review of Recent Advancements and Future Directions

作     者：Abdullah, Norizal Ayob, Masri Chun Lam, Meng Sabar, Nasser R. Kendall, Graham Khairulamirin Md Razali, Mohamad 

作者机构：Univ Kebangsaan Malaysia Fac Informat Sci & Technol Ctr Artificial Intelligence Technol CAIT Data Min & Optimizat Lab Bangi 43600 Selangor Malaysia Univ Kebangsaan Malaysia Fac Informat Sci & Technol Ctr Artificial Intelligence Technol CAIT Mixed Real & Pervas Comp Lab Bangi 43600 Selangor Malaysia La Trobe Univ Dept Comp Sci & Informat Technol Melbourne Vic 3086 Australia MILA Univ Sch Engn & Comp Nilai 71800 Negeri Sembilan Malaysia Univ Nottingham Sch Comp Sci Univ Pk Nottingham NG7 2RD Notts England Univ Kebangsaan Malaysia Fac Informat Sci & Technol Bangi 43600 Selangor Malaysia 

出 版 物：《IEEE ACCESS》 (IEEE Access)

年 卷 期：2025年第13卷

页      面：5203-5218页

核心收录：

基　　金：Ministry of Higher Education (MoHE) Malaysia  under Transdisciplinary Research Grant Scheme [TRGS/1/2019/UKM/01/4/1] 

主　　题：Medical services Optimization Market research Machine learning Databases Sustainable development Search problems Personnel Mathematical programming Physician scheduling personnel scheduling systematic literature review combinatorial optimization operational research sustainable development goals sustainable development goals 

摘      要：The Physician Scheduling Problem (PSP) has emerged as a critical challenge in healthcare management, directly relevant to Sustainable Development Goal 3 (SDG 3) - Good Health and Well-being. Driven by physician shortages, rising operational costs, and the need for efficient workforce planning, PSP affects the quality of patient care, staff satisfaction, and the overall efficiency of the healthcare system. While previous reviews have addressed PSP, they are lacking in a comprehensive analysis of recent optimization methodologies and their effectiveness. This work aims to bridge this gap by analyzing 60 research studies which addressed PSP, published between January 2014 and June 2024. Our study also extends the problem definition, constraints, evaluation functions, and the variants of PSP. We examine a wide range of optimization methodologies, including mathematical programming, heuristics, matheuristics, and machine learning, highlighting their strengths and limitations in addressing the multifaceted nature of PSP. This review also analyzes the datasets used in PSP research, noting the lack of standardized benchmarks. Key findings reveal the prevalence of mathematical optimization methods, the growing importance of multi-objective optimization and robustness, as well as the potential of machine learning and data-driven approaches. Future research directions are outlined, emphasizing the need for more scalable algorithms, real-time scheduling capabilities, improved user interfaces, and comprehensive validation studies. This review contributes to the advancement of PSP optimization, aiming to enhance healthcare workforce management, improve patient care, and ultimately address the pressing challenges faced by healthcare systems worldwide, thus supporting the achievement of SDG 3 and promoting universal health coverage.