Thermoregulation model JOS-3 with new open source code

A thermoregulation model, joint system thermoregulation model (JOS-3), was developed based on JOS-2. JOS-3 consists of 83 nodes, and human physiological responses and body temperatures are calculated using the backward difference method. In JOS-3, brown adipose tissue activity, aging effects, and heat gain by shortwave solar radiation at the skin, are installed to predict human physiological responses, considering personal characteristics in transient and non-uniform thermal environments. In addition, methods to determine shivering thermogenesis, sweating distribution, and basal metabolic rate were modified from those used in JOS-2. We coded JOS-3 in Python-3 and published the JOS-3 package. This model was validated by comparing its results with those of human subject tests conducted under stable and transient conditions. It was confirmed that JOS-3 has a higher accuracy for heat production in young and older subjects and mean skin temperature in older subjects than JOS-2 under cold environmental conditions. Moreover, JOS-3 was simulated in nine transient conditions. Consequently, the root mean-squared error (RMSE) of the rectal and mean skin temperatures between the predicted and experimental values were 0.12–0.38 and 0.58–0.83 °C, respectively.