Time series forecasting of bed occupancy in mental health facilities in India using machine learning

• Published in: Scientific reports Vol. 15; no. 1; pp. 2686 - 18
• Main Authors: Avinash, G., Pachori, Hariom, Sharma, Avinash, Mishra, SukhDev
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 21.01.2025; Nature Publishing Group; Nature Portfolio
• Subjects: 692/700/1538; 692/700/228; 692/700/3934; 692/700/478; Accuracy; Algorithms; Bed Occupancy - statistics & numerical data; Bed Occupancy - trends; Bed occupancy forecasting; Cognitive ability; COVID-19; Efficiency; Emergency medical care; Forecasting; Forecasting techniques; Health care; Health facilities; Health services; Healthcare resource management; Hospital systems; Hospitals, Psychiatric - statistics & numerical data; Humanities and Social Sciences; Humans; India; Learning algorithms; Machine Learning; Machine learning in healthcare; Mental disorders; Mental health; Mental health care; Mental health hospital; Mental Health Services - statistics & numerical data; Mental institutions; Monte Carlo simulation; multidisciplinary; Neural networks; Optimization techniques; Pandemics; Patient care optimization; Psychiatry; Queuing theory; Regression analysis; Resource allocation; Resource management; Science; Science (multidisciplinary); Time series; Time series analysis; India; Bed occupancy forecasting; Healthcare resource management; Patient care optimization; Mental health hospital; Time series analysis; Machine learning in healthcare
• DOI: 10.1038/s41598-025-86418-9
• ISSN: 2045-2322

Machine learning models are vital for forecasting and optimizing healthcare parameters, especially in the context of rising mental health issues in India and globally. With increasing demand for mental health services, effective resource management, like bed occupancy forecasting, is crucial to ensure proper patient care and reduce the burden on healthcare facilities. This study applies six machine learning models, namely Support Vector Regression, eXtreme Gradient Boosting, Random Forest, K-Nearest Neighbors, Gradient Boosting, and Decision Tree, to forecast weekly bed occupancy of the second largest mental hospital in India, using data from 2008 to 2024. Accuracy of models were evaluated using Mean Absolute Percentage Error, and Diebold–Mariano test for assessing differences in predictive performance. Further, we forecast the bed occupancy, providing crucial insights for healthcare administrators in capacity planning and resource allocation, supporting data-driven decisions and enhancing the quality of mental health services in India.