Does PHS Model Predict Acceptable Skin and Core Temperatures While Wearing Protective Clothing.

Research output: Contribution to conferencePaper, not in proceeding


Mathematical modeling is very important when experimental settings with human subjects are restricted to thermal limits necessary to protect the individual. The predicted heat strain (PHS) model has been published AS ISO 7933 for about six years. It describes a method for predicting the sweat rate and internal core temperature that the human body will develop in response to the working conditions. The PHS model was developed based on thousands of laboratory and field experiments collected from eight European laboratories. However, most of the laboratory and field tests were performed on human subjects with light clothing ensembles (0.38±0.34 clo < Icl < 0.77±0.18 clo). The prediction of physiological responses while human wearing highly insulating protective clothing might be weak. In order to check the prediction accuracy of current PHS model while using protective clothing, we conducted totally series of human subject tests at a simulated hot environment. The results of 18 tests involving the high visibility (HV), military (MIL) and firefighting (FIRE) clothing are reported here. Six human subjects were asked to walk on a treadmill at 4.5 km/h at 40 oC for 70 min. Two humidity levels were chosen: 2 kPa (RH = 27 %) and 3 kPa (RH = 41 %) depending on the garment. The rectal temperature, skin temperature, heart rate and metabolic rate were measured. The clothing and the subjects were weighed before and after the exposure in order to calculate the sweat and evaporation rate. The observed and predicted rectal temperatures and mean skin temperatures were compared. The PHS model failed to predict the final rectal temperature in FIRE and the predicted estimate was 1.83 oC higher than the observed value after 63-min exposure. The predicted curve showed a much deeper linear increase during the whole exercise. None of the predicted mean skin temperatures during the three testing scenarios were accurately predicted. The PHS model was consistently providing conservative mean skin temperature evaluations. The predicted curve in HV and MIL showed a much shallower increase during the early portion of the exposure and plateaued at temperatures lower than ever achieved by the subjects. The observed sweat rates were 556±110 g/h in HV, 717±200 g/h in MIL, and 834±274 g/h in FIRE. There was no significant difference between the predicted total sweat values and the experimental data (P=0.073). In summary, the PHS model produce prediction of core temperature which has an unacceptable error when human wore thick protective clothing. The weak prediction on the mean skin temperature in HV and MIL was in agreement with the empirical prediction equation in the source codes has the poorest and lowest correlation when a clothed human subject exercised at the humidity level above 2 kPa. It is therefore recommended that the PHS model should be amended to development and validated by manipulation of individual algorithms or physical (or physiological) parameters.


Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Production Engineering, Human Work Science and Ergonomics


  • PHS model, heat stress, skin temperature, protective clothing, core temperature
Original languageEnglish
Number of pages5
Publication statusPublished - 2010
Publication categoryResearch
Event8th International Thermal Manikin and Modeling Meeting (8I3M) - Victoria, BC, Canada
Duration: 2010 Aug 23 → …


Conference8th International Thermal Manikin and Modeling Meeting (8I3M)
CityVictoria, BC
Period2010/08/23 → …

Total downloads

No data available