Adaptive EWMA Control Chart by Adjusting the Risk Factors through Artificial Neural Network

Abstract

This article focuses on operationalizing a quality improvement framework to improve the quality and safety of patients in hospitals. Assessing the effectiveness of health care services, especially utilizing various patient health statuses, poses several difficulties. The artificial neural networks model assesses patient risk factors and enhances proper management. Our proposed approach extends the exponentially weighted moving average (EWMA) control chart to a risk‐adaptive EWMA chart. This chart is developed from residuals estimated from the artificial neural networks model, thus allowing an assessment of actual data from the patients undergoing cardiothoracic surgery. For patient evaluation, we employ artificial neural networks to establish the suggested control chart. The results indicate that this chart outperforms other methods, such as the risk‐adjusted EWMA chart for shift detection, which may present improvements in the healthcare system's patient care.