Simulation of patient flow in multiple healthcare units using process and data mining techniques for model...

Area of Science:

• Healthcare Systems Engineering
• Computational Health Informatics
• Simulation Modeling

Background:

• Accurate simulation of patient flow is crucial for optimizing healthcare operations and resource allocation.
• Existing simulation methods may lack the detail and realism needed for complex clinical pathways.
• Electronic Health Records (EHRs) contain rich data for improving simulation model accuracy.

Purpose of the Study:

• To develop and validate a hybrid simulation approach for patient flow modeling.
• To automate model identification using data-driven techniques.
• To enhance the realism and detail of patient flow simulations, specifically for acute coronary syndrome (ACS).

Main Methods:

• Combined data mining, text mining, process mining, and machine learning for EHR analysis.
• Utilized discrete-event simulation (DES) and queueing theory for patient flow simulation.
• Developed a hybrid model implemented in Python using SimPy and SciPy libraries.

Main Results:

• Successfully simulated patient flow for acute coronary syndrome (ACS) cases.
• Identified distinct clinical pathways (CPs) from EHR data to enhance simulation realism.
• Experimental study demonstrated improved accuracy in simulating patient length of stay.

Conclusions:

• The proposed hybrid simulation approach provides a robust framework for modeling complex patient flows.
• The methodology enables more realistic and detailed simulations for various healthcare applications, including decision-making and operational optimization.
• This approach offers a conceptual, methodological, and programming foundation for diverse simulation scenarios.