The Predicted Mean Vote (PMV) and Percentage of People Dissatisfied (PPD) evaluate thermal comfort conditions from the energy point of view. By analyzing the destroyed exergy and the exergy transferred to the environment, it is possible to evaluate the human thermal model with a higher degree of precision to different aspects of environmental parameters. Thus, phenomenological models were adapted to represent men (one dressed in a full suit and the other in lighter clothes) and women (in luteal phase and in follicular phase, with the same clothing pattern) from computational simulations to compare the thermodynamic behavior for different reference environments. The Cooling Load Temperature Difference method was used to model a building and evaluate the energy expenditure required for its cooling, considering the models of the human body as occupants. A vapor compression refrigeration cycle was modeled to evaluate the exergy destruction associated with the air conditioning system. The exergy-based indicators were compared with PMV and PPD. The thermal comfort points obtained using both methods are, 23.5/24 \(^\circ \) C for men in suits, 25/25 \(^o\) C for men in lighter clothing, 25.5/26 \(^\circ \) C for women in the follicular phase; and 24.8/25.5 \(^\circ \) C for women in the luteal phase. Women demand higher comfort temperatures than men during the follicular phase. During the luteal phase, the thermal comfort temperatures are close to that of men in light clothing and lower than for the follicular phase. We observed that the energy consumption decreases linearly with the increase in the operating temperature of the air conditioning and that the increase in the operating temperature from 23.5 \(^\circ \) C (close to the comfort temperature of men in suits) to 25.5 \(^\circ \) C (close to the comfort temperature observed in women and men with light clothing) implies a reduction of 4.1% of the cooling thermal load, representing a daily energy savings of 0.585 kWh.