Machine learning-based lean service quality improvement by reducing waiting time in the healthcare sector

Purpose The study presented healthcare service quality, lean thinking and Six Sigma to enhance patient satisfaction. Moreover, the notion of machine learning is combined with lean service quality to bring about the fundamental benefits of predicting patient waiting time and non-value-added activities to enhance patient satisfaction. Design/methodology/approach The study applied the define, measure, analyze, improve and control (DMAIC) method. In the define phase, patient expectation and perception were collected to measure service quality gaps, whereas in the measure phase, quality function deployment (QFD) was employed to measure the high-weighted score from the patient's voice. The root causes of the high weighted score were identified using a cause-and-effect diagram in the analysis phase. Findings The study employed a random forest, neural network and support vector machine to predict the healthcare patient waiting time to enhance patient satisfaction. Performance comparison metrics such as root-mean-square error (RMSE), mean absolute error (MAE) and R2 were accessed to identify the predictive model accuracy. From the three models, the prediction performance accuracy of the support vector machine model is better than that of the neural network and random forest models to predict the actual data. Practical implications Lean service quality improvement using DMAIC, QFD and machine learning techniques can be generalized to predict patient waiting times. This study provides better realistic insights into patient expectations by announcing waiting times to enable data-driven service quality deliveries. Originality/value Prior studies lack lean service quality, Six Sigma and waiting time prediction to reduce healthcare waste. This study proposes lean service quality improvement through lean Six Sigma (LSS), i.e. DMAIC and machine learning techniques, along with QFD and cause-and-effect diagram.