Research on obstetric ward planning combining lean thinking and mixed-integer programming.

Area of Science:

• Healthcare Management
• Operations Research
• Hospital Administration

Background:

• Existing research focuses on specific scheduling problems but lacks macroscopic approaches to ward allocation.
• There is a need for optimized obstetric ward planning to enhance patient care and operational efficiency.

Purpose of the Study:

• To refine and stratify obstetric wards for improved medical services for pregnant women.
• To optimize obstetrician and midwife work efficiency through strategic ward allocation.
• To model ward allocation as a mixed-integer programming problem to maximize patient flow.

Main Methods:

• Application of lean thinking to categorize obstetric wards into observation, cesarean section, and natural delivery types.
• Utilizing CPLEX to solve the mixed-integer programming model for ward allocation.
• Employing multivariate Generalized Linear Models (GLM) in R for analyzing factors influencing patient flow.

Main Results:

• CPLEX optimization yielded a 19-25% increase in maximum patient flow compared to unrefined methods.
• GLM regression analysis identified key positive and negative correlation factors impacting patient flow.

Conclusions:

• Stratifying obstetric wards enhances efficiency for healthcare professionals and improves patient services.
• Mixed-integer programming provides a macroscopic solution for optimizing obstetric hospital capacity.
• GLM analysis further supports capacity improvement by identifying influential factors.