Bonfring International Journal of Industrial Engineering and Management Science
Online ISSN: 2277-5056 | Print ISSN: 2250-1096 | Frequency: 4 Issues/Year
Impact Factor: 0.541 | International Scientific Indexing(ISI) calculate based on International Citation Report(ICR)
Explainable Machine Learning Framework for Predicting Hospital Length of Stay to Enhance Healthcare Resource Management
Abstract:
Hospitals all over the world are in more trouble than ever to manage the resources for healthcare: more and more patients are seen in the hospitals, hospital beds are not enough, and operational costs are increasing. Length of Stay (LOS) Patients is one of the critical factors affecting the efficiency of the hospital since it determines bed availability, treatment plan and general healthcare services. Conventional statistical methods and manual decision-making models are not always able to accurately predict LOS because of the complexity and the huge amount of data in electronic health records (EHRs). This shortcoming can result in the ineffective distribution of resources, higher waiting times, and poorer care of the patients. As a solution to this practical problem, research offers a healthcare analytics system with a predictive approach to Length of Stay in hospital based on machine learning and Explainable Artificial Intelligence (XAI). The system proposed the use of XGBoost and Long Short-Term Memory (LSTM) models to detect complicated trends in patient data. In addition, SHAP (Shapley Additive exPlanations) is also included to make the prediction process transparent and interpretable so that healthcare professionals can learn what factors contribute to LOS predictions. It also has interactive dashboards that enable the hospital administrator and patients to view insights into healthcare and track recovery progress. The presented solution can help optimise hospital resource efficiency, practices of clinical decision making and increase patient involvement. Finally, the system will help to achieve Sustainable Development Goal (SDG) 3: Good Health and Well-Being through ensuring effective healthcare provision and better access to high-quality medical care.
Keywords: Machine Learning, Hospital Length of Stay Prediction, Electronic Health Records (EHR), Explainable Artificial Intelligence (XAI), XGBoost, Long Short-Term Memory (LSTM), Healthcare Analytics.
Volume: 16 | Issue: 1
Pages: 5-12
Issue Date: March , 2026
DOI: 10.9756/BIJIEMS/V16I1/BIJ26002
 |
|