Thermal Comfort (PMV-PPD) Simulator Back
Building Physics

Thermal Comfort (PMV-PPD) Simulator

A tool for Fanger's thermal-comfort model (ISO 7730), which turns "hot" and "cold" into numbers. Adjust the six factors — air temperature, radiant temperature, air speed, humidity, metabolic rate and clothing — to see the Predicted Mean Vote PMV, the Predicted Percentage of Dissatisfied PPD and the operative temperature update in real time, and design comfortable HVAC for offices and homes.

Parameters
Air temperature t_a
°C
Dry-bulb temperature around the occupant
Mean radiant temperature t_r
°C
Average radiant temperature of walls, windows and ceiling
Air speed v
m/s
Average air velocity around the body
Relative humidity RH
%
Metabolic rate M
met
Activity level. Seated 1.0 / office work 1.2 / brisk walk 3.0
Clothing insulation I_cl
clo
Garment insulation. Summer 0.5 / winter suit 1.0 / heavy 1.5
Results
Predicted Mean Vote PMV
Dissatisfied PPD (%)
Thermal sensation
Clothing surface temp t_cl (°C)
Operative temp t_op (°C)
Comfort verdict
Six comfort factors — view around the body

The six factors that set the PMV (air temperature, radiant temperature, air speed, humidity, metabolic rate, clothing) are shown around the figure. The background bar and the figure colour track the thermal sensation (blue = cold / green = neutral / red = hot).

PMV vs air temperature t_a
PPD vs PMV
Theory & Key Formulas

$$PMV=\big(0.303\,e^{-0.036M}+0.028\big)\,L$$

Predicted Mean Vote PMV. M: metabolic rate [W/m²], L: the body's thermal load (heat production minus the six heat-loss terms) [W/m²]. The coefficient represents the dulling of sensation.

$$PPD=100-95\,e^{-(0.03353\,PMV^{4}+0.2179\,PMV^{2})}$$

Predicted Percentage of Dissatisfied PPD [%]. An even function of PMV with its minimum of 5% at PMV=0.

$$t_{op}=\frac{t_a+t_r}{2}$$

Operative temperature t_op [°C], the mean of air temperature t_a and mean radiant temperature t_r. Comfort means |PMV| < 0.5 (PPD < 10%).

What is Thermal Comfort (PMV)?

🙋
Even in the same room, some people say "it's hot" and others say "it's fine". Can that sensation really be predicted by calculation?
🎓
Good question. In the 1970s the Danish professor Fanger turned exactly that into an equation. It is the PMV — Predicted Mean Vote. Put many people in the same environment and it predicts what they will say on average, on a seven-point scale from -3 (cold) to +3 (hot), with 0 meaning "neutral, just right". The core idea is the body's heat balance: if the heat the body produces and the heat it loses are balanced you feel neutral, too much production feels hot, too much loss feels cold.
🙋
I see. But looking at the sliders on the left, there is much more than temperature — radiant temperature, clothing... isn't temperature enough?
🎓
That is the interesting part of PMV. Comfort is set by six factors. Four are environmental — air temperature, mean radiant temperature, air speed and humidity. Two belong to the body — metabolic rate and clothing. For example, near a window in winter the air may be 24°C, but body heat radiates away to the cold glass. So when the mean radiant temperature is low, you feel cold at the same air temperature. That is why looking only at the thermostat does not make a room comfortable.
🙋
PPD also appears in the results. How is it different from PMV?
🎓
PPD is the Predicted Percentage of Dissatisfied — what fraction of people feel "unhappy" with the environment. The interesting thing is that even when PMV is exactly 0, PPD never reaches zero; the minimum is 5%. No matter how hard you try, 1 in 20 will say "too hot" or "too cold". So the HVAC target is not "everyone satisfied" but "|PMV|<0.5, PPD<10%", i.e. fewer than one in ten dissatisfied. Move the PMV slider on the chart below and you will see the U-shaped curve with its floor at 0.
🙋
So when it's hot, is lowering the set temperature the only option? The chart suggests radiant temperature and air speed matter too.
🎓
Exactly. In summer, when it is a little hot, raising the air speed from 0.1 to about 0.6 m/s increases convective heat loss and lowers the PMV — that is why adding a fan works. In winter, insulating the windows raises the radiant temperature, so you can lower the air temperature and stay comfortable, which saves energy. Metabolic rate matters too: the optimum temperature for office work at 1.2 met and standing work at 2.0 met differs by 3-4°C. In the same office, "just right" really does differ between people who move around and those who sit all day.
🙋
One last thing. Changing the clothing value swings the PMV dramatically. How is that used in design?
🎓
"Cool Biz" and "Warm Biz" are exactly that. Assuming 0.5 clo in summer and 1.0 clo in winter gives the justification that PMV stays in the comfort band even at 28°C in summer and 20°C in winter. Conversely, if a workplace insists on suits all year and you raise the cooling set point, the larger clothing value swings PMV positive and dissatisfaction rises. PMV calculation is the common language for considering HVAC settings and dress codes together. Try moving the six factors one by one in this tool and feel how much each one matters.

Frequently Asked Questions

PMV (Predicted Mean Vote) predicts the average thermal sensation of a large group of people in the same environment on a seven-point scale from -3 (cold) through 0 (neutral) to +3 (hot). PPD (Predicted Percentage of Dissatisfied) is the percentage of people who feel the environment is too hot or too cold. PPD is a function of PMV: even at PMV=0 at least 5% of people are dissatisfied, and the comfort target is |PMV|<0.5, which corresponds to PPD below 10%.
PMV depends on six factors. Four are environmental — air temperature, mean radiant temperature, air speed and relative humidity — and two belong to the occupant: metabolic rate (met) and clothing insulation (clo). At the same air temperature you feel cold near a window where the radiant temperature is low, and warm when wearing more clothing. Fanger's model combines these six factors into a single heat-balance equation and derives PMV from the body's thermal load. HVAC design must control radiation, air movement and humidity, not just temperature.
Metabolic rate reflects activity: seated rest is 1.0 met, office work 1.2 met, light standing work 1.6-2.0 met and brisk walking around 3.0 met. One met equals 58.15 W/m² of body heat output. Clothing insulation is 0 clo (nude), about 0.5 clo for light summer clothing, 0.7 clo for a long-sleeve shirt and trousers, 1.0 clo for a winter suit and 1.5 clo or more for heavy winter wear. One clo equals 0.155 m²·K/W of insulation. PMV varies strongly with activity and clothing even in the same room.
PMV-PPD is a mean-value index for steady-state, air-conditioned interiors such as offices. Accuracy drops in three cases: (1) naturally ventilated spaces where occupants open windows and adjust clothing, (2) transient conditions where temperature changes quickly, and (3) local discomfort such as cold feet, overhead radiation or draught. It also ignores individual differences — even at PMV=0 about 5% are dissatisfied. In practice PMV is used alongside local comfort indices such as draught rate (DR) and vertical air temperature difference.

Real-World Applications

HVAC design for office buildings: Office building HVAC is designed so that PMV falls in the comfort band (|PMV|<0.5, PPD<10%). Designers assume a metabolic rate of 1.2 met (office work) and clothing of 0.5 clo in summer / 1.0 clo in winter, and adjust not only air temperature but the window insulation (radiant temperature) and supply-air velocity (air movement). Perimeter zones near windows tend to feel cold due to radiation, so designers add a separate HVAC loop for the perimeter zone or use Low-E glass to limit the drop in radiant temperature.

The basis of Cool Biz and Warm Biz: Set-point guidelines such as "28°C in summer, 20°C in winter" are backed by the calculation that, with clothing adjusted to 0.5 clo in summer and 1.0 clo in winter, PMV stays inside the comfort band. Put the other way round, moving the set point without changing the dress code pushes PMV out of the comfort band and dissatisfaction jumps. PMV calculation is the common language for discussing HVAC and clothing together.

Comfort assessment for homes and shops: In homes, underfloor heating and radiant panels raise the radiant temperature directly, so comfort is achieved at a lower air temperature, which saves energy. Conversely, over-cooling a summer shop swings PMV negative for lightly dressed customers, who then complain of cold. In spaces where customers and staff have different clothing and metabolic rates, checking PMV for both conditions is the practical approach.

Linking with building simulation and BEMS: Feeding the indoor temperature, radiation and air-velocity distribution computed by load-calculation software or CFD lets you evaluate PMV and PPD at each point in the space. Building energy management systems (BEMS) increasingly embed PMV control, using PMV rather than temperature as the control target. This tool is useful for intuitively checking the single-point PMV calculation that underpins all of that.

Common Misconceptions and Pitfalls

The most common misconception is that "controlling air temperature alone makes a room comfortable". Of the six factors that set PMV, air temperature is just one. Near a window in winter the air may be 24°C, yet body heat radiates to the cold glass surface, the mean radiant temperature drops and you feel cold. In summer, radiation from sun-heated walls and ceiling can make a room feel hot even after the set point is lowered. Do not trust the thermostat number alone — look at the environment including radiation, air movement and humidity.

Next, the belief that "if PMV=0 everyone is satisfied". As the PPD equation shows, even at the optimum point where PMV is exactly 0, PPD is 5% — 1 in 20 people are dissatisfied. This comes from individual differences (basal metabolism, acclimatisation, preference) and cannot be designed away. So the HVAC target is not "everyone satisfied" but "keep dissatisfaction below 10% (|PMV|<0.5)". Understand that "complaints never reach zero" is not a design fault but a fundamental limit of the thermal environment.

Finally, that "PMV is a universal index usable in any space" is not true. PMV-PPD was built for a uniform, air-conditioned interior in steady state, such as an office. In a naturally ventilated space where occupants open windows and change clothing, people adapt actively, so PMV overestimates dissatisfaction compared with reality (an adaptive model is used in that case). Local discomfort such as only the feet being cold or only the head receiving strong radiation does not show up in PMV. In practice PMV is used alongside local comfort indices such as draught rate (DR) and vertical temperature difference, and treated as a "whole-body average" guide.

How to Use

  1. Enter air temperature (18–26°C typical range) and mean radiant temperature in the top fields
  2. Set air velocity (0.1–1.0 m/s) and relative humidity (30–70%) in the middle controls
  3. Input clothing insulation (0.5–2.0 clo for business to winter wear) and metabolic rate (1.0–2.0 Met for seated to light activity)
  4. Click simulate; the tool computes PMV using ISO 7730 Fanger equations and displays thermal sensation category and PPD percentage

Worked Example

Office space: air temperature 22°C, mean radiant temperature 21°C, air velocity 0.3 m/s, relative humidity 45%, clothing 1.0 clo (business casual), metabolic rate 1.2 Met (seated desk work). Result: PMV = −0.18 (slightly cool), PPD = 5.2% dissatisfied, operative temperature 21.5°C, clothing surface temperature 28.3°C. Verdict: acceptable comfort per ASHRAE Standard 55.

Practical Notes

  1. PMV scale ranges −3 to +3; target −0.5 to +0.5 for optimal comfort zones minimizing dissatisfaction below 10% PPD
  2. Radiant asymmetry (sun-facing glass, cold walls) can increase PPD by 15–25%; use adjusted mean radiant temperature if needed
  3. High humidity (70%+) at warm temperatures (24°C+) triggers rapid PPD rise; dehumidification is cost-effective in humid climates
  4. Seasonal clothing adjustments critical: winter 1.5 clo versus summer 0.6 clo shifts comfort baseline by 2–3°C; retrain occupants accordingly