Abstract:In the coronavirus epidemic, many Chinese hospitals have established buffer zones to prevent the spread and transmission of the virus. The buffer zone is a monitored and separate area where the patients who need hospitalizations after the quick treatments in the emergency department can temporarily wait for the Covid-19 test and receive some healthcare services to stabilize their conditions. Because the beds in the buffer zones are limited, the managers face the patient admission control problem for the buffer zone. This management and control problem is challenging since the patient arrivals are uncertain, and the patients’ conditions are different. In this paper, we build the infinite- and finite-horizon Markov decision process (MDP) models for this problem. We use the uniformization method to discretize the patient flow. We propose various iteration algorithms to solve the MDP models and obtain the optimal and threshold policies. Numerical experiments validate the advantages of the policies obtained by the algorithms in this paper over the current policies of hospitals. Note to Practitioners—The ongoing COVID-19 pandemic has been causing enormous damage to people’s health, jobs, and well-being. COVID-19 has affected almost all countries globally and has changed the operation mode of the healthcare system, especially the hospitals. The hospitals are the frontlines of healthcare service and the battle with the COVID-19 pandemic. This article is motivated by our collaborations with hospitals in Shanghai, China. In China, many hospitals establish buffer zones: a monitored area where the patients who need hospitalizations after the quick treatments in the emergency department can temporarily wait for the Covid-19 test and receive some healthcare services to stabilize their conditions. Because the zone’s capacity is limited, the managers must make dynamic patient admission control decisions according to multiple factors, such as patients’ health conditions and the usage of beds in the zone. We propose two MDP models to solve this complex problem. Several iteration algorithms are designed to solve the MDP models and obtain the optimal and threshold policies. Based on hospitals’ real-life data, we show the methods presented in this paper can help hospital managers make more reasonable decisions. Although we focus on the hospital’s buffer zone in China, the methodology and approach for this problem can be extended to other practical hospital management scenarios in the coronavirus pandemic. For example, For example, some hospitals have admission control problems for coronavirus patients due to hospital capacity limitations. The hospital has to decide if a patient is accepted as an inpatient or suggested to home quarantine. In such a case, the admission control problem can also be solved by the methodologies in the paper.

Hospitalization admission control of emergency patients using Markovian Decision Processes and discrete event simulation

Admission Control of Multi-Level Emergency Patients with Time-Varying Demands

Scalable approximate policies for Markov decision process models of hospital elective admissions

A Markov Decision Process model for elective inpatient admission with delay announcement

Real-time Scheduling of Semi-Urgent Patients under Waiting Time Targets

Admission Control in A Pure Loss Healthcare Network: Mdp and Des Approach

Admission Control of Hospitalization with Patient Gender by Using Markov Decision Process

Dynamic Patient Admission Control with Time-Varying and Uncertain Demands in COVID-19 Pandemic

Mdp Based Inpatient Premature Discharge Decision Making In Hierarchical Medical System

Markov Decision Process and Approximate Dynamic Programming for a Patient Assignment Scheduling problem

Scheduling inpatient admission under high demand of emergency patients

Design and Optimization Methods for Elective Hospital Admissions

An approximate dynamic programming approach to the admission control of elective patients

Dynamic Multi-Type Patient Advance Scheduling for a Diagnostic Facility Considering Heterogeneous Waiting Time Targets and Equity

A Markov Decision Process Model of Allocating Emergency Medical Resource among Multi-Priority Injuries.

Optimal Admission and Queuing Control with Reneging Behavior under Premature Discharge Decisions

On the decision support model for the patient admission scheduling problem with random arrivals and departures: A solution approach

A Discrete Event Simulation Based Approach for Admission Control of Pregnant Women in Obstetrics Units

MDP-based appointment scheduling for one examination of two types of outpatients

Optimal Dynamic Outpatient Scheduling for a Diagnostic Facility with Two Waiting Time Targets

Appointment Scheduling of Diagnostic Facilities Subject to Non-Stationary Emergency Demand and Waiting Time Targets