Critical Analysis of Machine Learning and Deep Learning Models for Predicting Medical Equipment Demand in Hospitals

Sureshbabu, Raghul (2025) Critical Analysis of Machine Learning and Deep Learning Models for Predicting Medical Equipment Demand in Hospitals. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (785kB) | Preview
Preview	PDF (Configuration Manual) Download (705kB) | Preview

Abstract

Accurate forecasting of medical equipment demand is a major challenge in hospital management. This paper addresses this gap by conducting a critical analysis of statistical, machine learning, and deep learning models for forecasting Medicare Durable Medical Equipment (DME) demand from 2014 to 2022. This analysis incorporates SARIMAX as a statistical baseline, and other algorithms such as XGBoost, Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and N-BEATS in order to predict the complex temporal dependencies with SHAP based explainability for interpretation.

In the primary experiments, XGBoost achieves the highest accuracy (R2 = 0.9812) and N-BEATS produces a similar result (R2 = 0.9801), where both clearly outperform other models. An additional experiment with recently released 2023 data confirms that XGBoost continued to show strong predictive accuracy (R2 = 0.9883), while N-BEATS performance decreased to 97% indicating that gradient boosting is more accurate and stable than other algorithms. This study is useful for the healthcare operations area by recommending a data-driven and interpretable forecasting approach that can assist in trustworthy and ethically consistent decision-making in the planning and procurement of medical equipment in hospitals.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Sahni, Vikas UNSPECIFIED
Subjects:	R Medicine > Healthcare Industry Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	09 Sep 2026 10:24
Last Modified:	09 Sep 2026 10:24
URI:	https://norma.ncirl.ie/id/eprint/9919

Actions (login required)

View Item