Defining Thermal Comfort

Each of the buildings achieves a different pattern of internal temperature over the design day. We need to be able to define what temperatures are acceptable and what are not. Inevitably there will be some difference between the comfort levels of individuals but we need to use some measure which will say, for example, 'most people will be acceptably comfortable provided temperatures do not go above or below such and such a range of temperatures'.

As peoples comfort is a function of expectation, we can assess results based on a medium term experience of climate. Thermal expectations will be higher in summer than winter and thus temperatures required to achieve comfort will be higher in summer.

This idea can be extended to a different location where a different experience is realised

We can thus define acceptable comfort ranges for different locations. However it is useful to have a universal guideline which enables us to apply the same principles to any location.

Michael Humphreys developed a rule which uses a monthly mean external temperature to determine an optimum temperature, referred to as Thermal Neutrality. This represents a state at which the maximum number of people feel neither hot nor cold. His rule was determined from investigations in a number of locations of buildings which were not mechanically cooled, many of which also were not heated.

The Temperature of thermal neutrality according to Humphreys is:

Tn = 11.9 + 0.534 * TavTav > 13oC
Tn = 19Tav < 13oC

Where Tav is the outdoor average monthly mean daily temperature

Whilst Thermal Neutrality might represent the optimum temperature, we may consider that if the temperature is within a certain range of this temperature we can still achieve comfort for the majority of people. Szokolay defines this range as between 2oC less than Thermal Neutrality to 2oC more than Thermal Neutrality ie a band 4oC wide centred on The Temperature of Thermal Neutrality

Whilst Humphries rule is universal, some have attempted to find rules for a more specific area. Auliciems has studied comfort in buildings in Australia and determined rules specifically for that country. His most simple rule takes a similar form to Humphries

Tn = 17.7 + 0.31 * Tav18.5 <= Tav <= 28.5

His particular area of interest was in adjusting air-conditioning controls, making them adaptive, depending on the outside climate. This may explain why his rule is slightly less dependent upon the outside temperature.

Auliciems also derived a rule which uses the average indoor temperature to determine Thermal Neutrality.


Humphreys, 1978: ‘Outdoor temperatures and comfort indoors’ BRS (BRE) Current Paper 53/78

Auliciems, 1989: 'AirConditioning in Australia III Thermobile Controls, Architectural Science Review, Vol 33 pp43-48

Szokolay, S. V. (1995): Thermal Design of buildings, RAIA Publications, Canberra