Elderly people are vulnerable to cold environments, due to aging’s impact on the density of nerve fibers. Nevertheless, automatic controls of HVAC systems do not consider the user’s real-time thermal sensation and most of algorithms were tested in manikins using climatic chambers or simulated by CFD. Indeed, a standardization of a non-invasive technique as infrared thermography (IRT) for human body responses was not detected in the literature review. This paper proposes a method to determine the indoor thermal comfort of elderly people by IRT. The campaigns took place in a nursing home of the Mediterranean Climate during Winter. A total of 15 old adults were monitored to measure the skin temperature of four face points (nose, forehead, cheekbone and chin) and clothing temperature. A thermal sensation questionnaire and indoor conditions were also collected. The results demonstrated that the IRT model could estimate the thermal exchange from three body-segments (head, back-pelvis, thorax-limbs) to the surroundings by user’s features and environmental parameters. The thermal neutrality was achieved when all facial points reached a skin temperature of 35ºC, the clothing temperature was equal to 31ºC, and the operative temperature and relative humidity were 23.5ºC and 54% respectively. Statistically, 31.70% of the variance of PMV could be attributed to increases of facial skin temperature, while changes in user’s features could have minor influence (from 0.88 to 6.40%). As regards the main contributions of heat losses, they were given by respiration (11-21%), convection (20-32%) and radiation (35-39%). Finally, an innovative HVAC control strategy was posed.
