Analysis of Saturation in the Emergency Department: A Data-Driven Queuing Model Using Machine Learning

Abstract

Emergency department is a key component of the health system where the management of crowding situations is crucial to the well-being of patients. This study proposes a new machine learning methodology and a queuing network model to measure and optimize crowding through a congestion indicator, which indicates a real-time level saturation.

Estimating emergency department crowding with stochastic population models

“…Modeling crowding has been a major challenge in disciplines such as operations research, marketing, transportation research, psychology, and healthcare management [4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21].…”

“…A considerable body of research focused on describing the average behavior in the population (e.g., arrival rate, length of stay, probability of making a specific decision based on crowding etc.). These models typically implemented methodologies such as queuing models [11,12], econometric analysis [4][5][6][7][8][9], time series analysis [10] (Kadri, Harrou, Chaabane, & Tahon, 2014), dynamic choice models [13,14], and performance analysis [16] (Chan, Farias, Bambos, & Escobar, 2012). While these methods are powerful for capturing average behaviors, they are less suitable for analyzing the fluctuating nature of crowding.…”

“…ED crowding has been intensively studied [10][11][12][13]15,16,[18][19][20][21][22]28,29], exploring a variety of operational metrics [27], and using variety of modeling techniques (see Hu, Barnes and Golden 2018 [20], Vanbrabant et al 2019 [27], and Sinreich and Marmor 2005 [25] for review). Our model offers a fourfold contribution: First, it captures average behavior, as well as spikes and events of overcrowding, which received less research attention [10][11][12][13]. Second, our model is compact.…”