ABSTRACT

This research investigated the key factors that influenced patients’ individual thermal sensations in a rehabilitation ward. Maintaining thermal comfort is important for occupant's well-being in healthcare facilities. The commonly used Predicted Mean Vote (PMV) thermal comfort model has limitations on considering an individual's needs, especially if the individual has impaired health. There was a lack of thermal sensation studies in medical settings. This study conducted a ten-week fieldwork in a real rehabilitation environment in order to develop a thermal sensation analysis model that could help understand individual patient's thermal needs. Traditional statistical models and artificial neural network (ANN)-based models, using real-world data including spatial and healthcare-related parameters, were established for a comparative study.

The results of the study unveiled the substantial influence of spatial and healthcare-related parameters on inpatients’ indoor thermal sensations. Furthermore, the ANN-based model demonstrated better performance in aligning with real-world conditions and in providing more accurate prediction outcomes compared to the traditional statistical model. These findings can be used by hospital designers and engineers to optimize the overall quality of the thermal environment within a healthcare environment.