Thermal comfort in buildings
Abstract:
The basic principles of creating environments for thermal comfort are well understood. This chapter presents those principles and uses them to address the new challenge of specifying conditions for sustainable thermal comfort. Currently accepted methods and Standards are presented which include the predicted mean vote (PMV) and the predicted percentage dissatisfied (PPD) thermal indices. A description of behavioural models leads to the equivalent clothing index (IEQUIV) and a suggestion that it is not necessary to heat offices above 20 °C nor cool them below 25 °C. If this practice were followed it would save significant resources and make a step change in progress towards sustainable thermal comfort. It is concluded that while much can be done, our philosophical, technical and scientific understanding of sustainable thermal comfort is still in its infancy.
5.1 Introduction
There are many ways of creating indoor environments that provide thermal comfort and the factors involved are generally well understood. The reduction of air temperature in an office during a heat wave using air conditioning systems can provide thermal comfort for the people in that office. The increase in air movement using fans or natural air flows, however, can also provide comfort even if the air temperature is not reduced. The conditions that provide thermal comfort are well understood, from a century of systematic laboratory and field research, involving studies in psychology, psychophysics, biophysics and physiology. However, the relationships among designing for thermal comfort and energy use, human behaviour, and sustainability have not been studied extensively and are not well understood. The principles for creating thermal comfort are a prerequisite to sustainable design. Use of the principles to achieve requirements for energy and sustainability is a subject of on-going investigation. Both the principles of thermal comfort and the practical design for thermal comfort in sustainable environments are addressed in this chapter.
5.2 Thermal comfort
Thermal comfort is a term that is generally regarded as a desirable or positive state of a person. It is used in relation to how warm or cold a person feels and is clearly related to the environment a person occupies. There are many levels of discussion regarding the meaning and nature of thermal comfort and there was much activity and debate in the 1960s and 1970s on this topic (see McIntyre, 1980, for a review). Much has been achieved since 1970 in understanding the conditions that create thermal comfort. However, because of the need to take our understanding to a higher level in order to meet increased and new requirements, these discussions are again coming to the fore. A starting point is the generally accepted basic definition of thermal comfort from ASHRAE (1966) and now adopted internationally: ‘Thermal comfort is that condition of mind which expresses satisfaction with the thermal environment’.
The emphasis is on a state of mind clearly linking thermal comfort with a ‘psychological’ response. So how do we know what state of mind a person is in? The obvious answer is that we ask them. Subjective scales for use in measuring thermal comfort are provided in Section 5.3. For most cases the definition above is acceptable. There is debate, however, about whether it is entirely correct. What if the state of mind expresses satisfaction but it is clear that the conditions could not be regarded as comfortable or even acceptable? If a person is not familiar with the concept of thermal comfort then he or she may not be disposed to report discomfort. Satisfaction may be expressed because dissatisfaction is not the state of mind. This could apply even where a person is sweating, sticky, hot and so on; a state of mind which expresses satisfaction but surely the person is not in thermal comfort. This should be distinguished from expression of satisfaction and hence implied thermal comfort due to expectation. The state of mind is then ‘uncomfortable’ but satisfaction is expressed because the person is not used to, nor expects, ‘any better’.
A further group of people where it is difficult to interpret the definition is for those who may have a limited concept of comfort or an inability to express it. Babies, young children, people with mental disabilities and others, would generally not be able to express a condition of mind. An alternative definition may then apply as it is clear (at least reasonable to assume) that all people in thermal environments would regard certain conditions as undesirable and negative and others desirable and positive (i.e. whatever their condition of mind).
The issue of whether the concept of thermal comfort is an international phenomenon is of interest if we are to specify optimal conditions for indoor environments worldwide. If the concept exists how is it expressed in terms of the culture and language of the population? It has generally been considered that the concept of thermal comfort does exist internationally and that conditions for thermal comfort can, for practical purposes, be assumed to be identical across the world. Arguments against this assumption have generally pointed to the likelihood, or natural expectation, that people in hot conditions would prefer ‘warm’ or higher temperature conditions for thermal comfort than those that live in temperate or cold climates. In fact, evidence suggests that this is not the case and that although people who live in hot or cold climates are better able to survive heat or cold respectively, conditions for thermal comfort do not vary (Fanger, 1970; Parsons, 2003).
In field studies of tropical climates it is often concluded that thermal comfort conditions are at higher temperatures than those for people who live in temperate climates. Supporting evidence often cites indoor temperatures which exist in buildings in tropical countries and assume that they are comfort temperatures. First, there is no reason why we can assume that these are comfort temperatures and second, even if people say they are satisfied, it may be for one of the reasons outlined above and probably if ‘actual’ comfort conditions were presented then this would be preferred.
A less neutral and more positive concept related to thermal comfort is thermal pleasure. A contributor to good feng shui is a refreshing wind providing a stimulating and desirable environment. Thermal pleasure can also be found in cool temperatures compensated for by the heat of the sun. Such conditions are usually beyond thermal comfort. Thermal pleasure can also be found where a person who is cold moves to a warmer environment and a person who is hot moves to a cooler environment. This is a transient, short-lived phenomenon and after a few minutes pleasure will disappear as the person adapts to the new conditions. Designing for thermal pleasure therefore is a function of the interaction of the thermal environment with other factors and the recognition that environments are dynamic.
5.3 Measurement of thermal comfort
One of the best ways of determining whether a group of people are comfortable is to ask them. Thermal sensation, comfort, pleasure, pain, as well as behavioural responses, are all psychological phenomena. There have been many useful studies to ‘correlate’ physical conditions and physiological responses with psychological responses. However, no model provides a more accurate prediction than measuring psychological responses directly.
Methods for measuring psychological responses range from psychophysical techniques (method of limits, method of magnitude estimation, multi-dimensional scaling, etc. – see Guildford, 1954) and simple (seven-point) scales often used in laboratory experiments, to the integration of techniques into questionnaires for practical surveys as well as behavioural measures.
5.3.1 Subjective measures
There are a number of subjective scales which have been used in the assessment of thermal environments; the most common of these are the seven-point scales of Bedford (1936) and ASHRAE (1966) (see Table 5.1).
The form and method of administering the scales is important. For example, a continuous form of the scale would be to draw a line through all points. This would allow subjects to choose points between ratings (e.g. between cool and cold, a rating of 1.6 on the ASHRAE scale, for example). In analysis of results this would enable parametric statistics to be used. However, maybe the investigator does not consider data ‘strong enough’ for this and is prepared only to use ordinal data (ranks) and non-parametric statistics. These and other points are of importance and for further information the reader is referred to a text on the design and analysis of surveys – e.g. Moser and Kalton (1971) – and on the use of subjective assessment methods – e.g. Sinclair (1990) as well as ISO DIS 28802 (2009) which describes scales for use in an environmental survey.
The ‘psychological’ interaction when the scale is administered may also influence results. Subjects are usually given the scale and asked to tick the place which represents ‘how they feel now’, for example. It is important to avoid ambiguity which may lead to a person providing their own interpretation, e.g. what the environment is generally like, or how other people may perceive it, etc. Other issues include range effects – the range provided, e.g. hot to cold, influences the judgement – and leading questions ‘you are uncomfortable aren’t you?’. Sinclair (1990) identifies the following important issues to be considered when constructing questionnaires: question specificity, language, clarity, leading questions, prestige bias, embarrassing questions, hypothetical questions and impersonal questions. Other issues include whether knowledge of results is given – for example, if responses are requested over time, is the subject informed of previous ratings he made? – and whether the ratings are given in the presence of others.
Investigations involving subjective measures therefore must be carefully planned. It should be emphasized that although there are many pitfalls, most can be relatively easily overcome and the use of simple subjective methods allows easy collection of important data, which can prove invaluable in the measurement of psychological responses.
ISO 10551 (1995) presents the principles and methodology for the construction and use of scales for assessing the environment. Scales are divided into two types: personal and environmental. Those related to the personal thermal state may be perceptual – how do you feel now? (e.g. hot), affective – how do you find it? (e.g. comfortable) and preference – how would you prefer to be? (e.g. warmer). Those related to the environment fall into two types: acceptance (e.g. is the environment acceptable?) and tolerance (e.g. is the environment tolerable?) An interesting point for an international standard is translation between languages, since in French, for example, one cannot easily use together ‘warm’, ‘hot’ and ‘very hot’. The fundamental principles and psychological phenomena, however, apply over all nationalities although language and cultural difference (in some cultures subjects may be reluctant to express dissatisfaction) will be important.
The selection of subjective scales will depend upon the population under investigation and an initial investigation may be necessary to identify meaningful dimensions. For example, in the investigation of the thermal comfort of clothing (Hollies et al., 1979), the seven-point thermal sensation scales are in general use; however, scales of stickiness, wetness, etc., are used for specific applications. The construction and use of simple questionnaires used in a thermal survey are given in Parsons (2003) and ISO DIS 28802 (2009) (Environmental survey).
5.3.2 Behavioural and observational measures
Thermal environments can affect the behaviour of individuals (move about, curl up, put on or take off clothing, become aggressive or quiet, change thermostat setting, etc.). This behaviour can be observed and aspects of it quantified and hence measured. A number of studies have observed the behaviour of householders in controlling internal temperatures by measuring temperatures in the homes (e.g. Weston 1951; Humphreys, 1978). The causes are not always clear. They may be related to increased heating costs to the preference for wearing lighter clothing, for example. More direct observations of behaviour have been made on schoolchildren using time-lapse photography and two-way mirrors. For these methods observer interference is of great importance and should be carefully considered (Humphreys, 1972; Wyon and Halmberg, 1972). Other behavioural measures could include the occupational density of a room (where there is choice) or a measure of accidents or critical incidents.
Drury (1990) provides a description of direct observational methods that he says have a high degree of face validity but a low degree of experimental control. For example, even if behavioural measures are correlated with thermal conditions, it cannot be concluded that thermal conditions are the (sole) cause. If the establishment of causality is not important, however, then observation of behaviour provides a useful measure. Behavioural measures can provide a method of observing both quantitative and qualitative ‘measures’ with little interference with what is being observed.
5.4 The thermal index: an assessment technique
A useful tool for describing, designing and assessing thermal environments is the thermal index. The principle is that factors that influence human response to thermal environments are integrated to provide a single index value. The aim is that the single value varies as human response varies and can be used to predict the effects of the environment. A thermal comfort index, for example, would provide a single number that is related to the thermal comfort of the occupants of an environment. It may be that two different thermal environments (i.e. with different combinations of various factors such as air temperature, air velocity, humidity and activity of the occupants) have the same thermal comfort index value. Although they are different environments, for an ideal index, identical index values would produce identical thermal comfort responses of the occupants. Hence environments can be designed and compared using the comfort index.
A useful idea is that of the standard environment. Here the thermal index is the temperature of a standard environment that would provide the ‘equivalent effect’ on a subject as would the actual environment. Methods of determining equivalent effect have been developed. One of the first indices using this approach was the effective temperature (ET) index (Houghton and Yagloglou, 1923). The ET index was, in effect, the temperature of a standard environment – air temperature equal to radiant temperature, still air, 100% relative humidity for the activity and clothing of interest – which would provide the same sensation of warmth or cold felt by the human body as would the actual environment under consideration.
5.5 Thermal comfort indices
5.5.1 The six basic ‘parameters’
It is generally agreed that any specification of thermal comfort conditions must consider the six basic parameters (variables). These are air temperature, radiant temperature, humidity, air velocity (i.e. the four environmental factors) and the clothing insulation and heat produced by the activity of a person (i.e. the two personal factors). An important point is that any and all of the six factors can influence thermal comfort and that it is the integrated influence of all of the six factors that determines thermal comfort response. ISO 7726 (1985) provides information concerning specification of instruments for the measurement of the environmental factors. ISO 8996 (1990) provides methods for the estimation of heat production by a person for different activities. 1 Met is defined as the heat produced by a sedentary person and is given the value of 58 Watts produced for every square metre of the body surface area (i.e.1.0 Met = 58 Wm− 2. Higher activity levels have higher values. ISO 9920 (1992) provides information concerning the thermal properties of clothing. For dry insulation a value of 1 Clo is defined as providing an insulation of 0.155 m2 K W− 1 (1.0 Clo often regarded as the insulation of a typical business suit; 0 Clo is for a naked person, 0.6 Clo light clothing and so on).
5.5.2 Fanger (1970)
The most significant landmark in thermal comfort research and practice was the publication of the book Thermal Comfort by Fanger (1970), which outlines the conditions necessary for thermal comfort and methods and principles for evaluating and analysing thermal environments with respect to thermal comfort. Fanger considered that existing knowledge of thermal comfort was inadequate and unsuitable for practical application, and his book is based upon research undertaken at the Technical University of Denmark and at Kansas State University, USA. The methods that he developed are now the most influential and widely used throughout the world. The reason for this success has been the consideration of the ‘user requirements’. He had the vision to recognize that it is the combined thermal effect of all (six basic parameters) physical factors which determines human thermal comfort, and that a practical method was required which could predict conditions for ‘average thermal comfort’ and consequences (in terms of thermal discomfort, e.g. percentage of people dissatisfied) of exposure to conditions away from those for ‘average thermal comfort’.
5.5.3 The comfort equation
Fanger (1970) defines three conditions for a person to be in (whole-body) thermal comfort:
A fourth condition is the absence of local thermal discomfort (e.g. caused by draught). The objective was to produce a comfort equation requiring input of only the six basic parameters and based on the above three conditions, to calculate conditions for thermal comfort. This was achieved using a rational analysis of heat transfer between the clothed body and the environment and experimental research. Fanger’s original work was not in SI units so the SI version presented below is taken from Olesen et al. (1982), ASHRAE (1989) and ISO 7730 (2005).
5.5.4 Heat balance
Fanger’s conceptual heat balance equation is:
where a description of terms used is given in Table 5.2.
Table 5.2
Terms used in the predicted mean vote (PMV)
| H | = | Internal heat production in the human body |
| Ed | = | Heat loss by water vapour diffusion through skin |
| Esw | = | Heat loss by evaporation of sweat from skin surface |
| Ere | = | Latent respiration heat loss |
| L | = | Dry respiration heat loss |
| K | = | Heat transfer from skin to outer surface of clothing |
| R | = | Heat transfer by radiation from clothing surface |
| C | = | Heat transfer by convection from clothing surface |
Source: Comfort equation of Fanger (1970)
Heat is generated in the body and lost at the skin and from the lungs. It is transferred through clothing where it is lost to the environment. Logical considerations, reasonable assumptions, and a literature review provide equations for each of the terms such that they can be calculated from the six basic parameters: air temperature (ta); mean radiant temperature (tr); partial vapour pressure (Pa pv); air velocity (v); clothing insulation (Icl); and metabolic heat production (M–W, where M is metabolic rate and W is external work). Table 5.3 provides the equations for the components of the heat balance equation to determine the PMV thermal comfort equations. tcl is the mean surface temperature of clothing and exposed skin; ts is mean skin temperature and fcl is a non-dimensional term for the ratio of the surface area of the clothed person to the surface area of the naked person; hc is the convective heat transfer coefficient in Wm− 2 K− 1.
Table 5.3
Equations for components of the heat balance equation used by Fanger (1970) in determining the PMV thermal comfort equations (see Table 5.2 for terms)
Note: Units for all components: W m− 2; Pa in Pascals; temperatures in °C and lcl in m2 KW− 1.
5.5.5 Sweat rate and skin temperature for comfort
Heat balance is a necessary but not sufficient condition for comfort. The body can be in heat balance but uncomfortably hot due to sweating or uncomfortably cold due to vasoconstriction and low skin temperatures. Skin temperatures and sweat rates required for comfort tsk,req, and Ersw,req, depend upon activity level. Rohles and Nevins (1971) provide the following equations:
By substituting tsk,req and Ersw,req terms into the heat balance equation, the method of combination of the six basic parameters which produce thermal comfort can be expressed in a comfort equation (see Fanger, 1970; Parsons, 2003).
5.5.6 Predicted mean vote (PMV) and predicted percentage dissatisfied (PPD)
To provide a method for evaluating and analysing thermal environments, Fanger made the proposal that the degree of discomfort will depend on the thermal load (L). This he defined as ‘the difference between the internal heat production and the heat loss to the actual environment for a man hypothetically kept at the comfort values of the mean skin temperature and the sweat secretion at the actual activity level’. In comfort conditions the thermal load will be zero. For deviations from comfort the thermal sensation experienced will be a function of the thermal load and the activity level. For sedentary activity, Nevins et al. (1966) and Fanger (1970) provide data and McNall et al. (1968) provide data for four activity levels (from 1396 subjects exposed for 3 hours in 0.6 Clo KSU uniform). This provided an equation for the predicted mean vote (PMV) of a large group of subjects if they had rated their thermal sensation in that environment on the following scale:
| sensation | PMV |
| Hot | + 3 |
| Warm | + 2 |
| slightly warm | + 1 |
| Neutral | 0 |
| slightly cool | − 1 |
| Cool | − 2 |
| Cold | − 3 |
where Icl values are in units of Clo.
Fanger (1970) presents tables showing PMV values for 3500 combinations of the variables. These are now unnecessary as the calculations of PMV can easily be made on a personal computer; see ISO 7730 (2005) and Parsons (1993).
The predicted percentage of dissatisfied (PPD) provides practical information concerning the number of potential complainers. The data of nevins et al. (1966), Rohles (1970) and Fanger (1970) provided a relationship between the percentage of dissatisfied and the mean comfort vote (see Fig. 5.1 and ISO 7730, 2005):
Fanger (1970) describes a method for the use of the PMV and PPD in practical applications and includes examples involving an analysis of a large room and the use of a thermal non-uniformity index, lowest possible percentage of dissatisfied (LPPD), indicating the ‘best’ that can be achieved by changing the average PMV value in the room only – by changing average air temperature, for example – and hence indicating where specific areas of the room require attention.
Data to allow the determination of some PMV and PPD values are provided in Tables 5.4, 5.5, and 5.6.
Table 5.4
Examples of estimates of clothing insulation values (Icl) for use in the PMV thermal equation of Fanger (1970)
Table 5.5
Examples of estimates of metabolic rates (M) for use in the PMV thermal comfort equation of Fanger (1970)
Table 5.6
Predicted mean vote (PMV) values from Fanger (1970). Assume RH = 50%; still air and ta = tr. PMV: + 3, hot; + 2, slightly warm; + 1, warm; 0, neutral; − 1, slightly cool; − 2, cool; − 3, cold
5.6 International Standards and thermal comfort
Optimum indoor air temperatures have often been the subject of debate, and suggested limit or guidance values for buildings have been proposed over many years, by a number of professional institutions and in legislation. There is an increasing international interest in providing guidance to ensure protection and good practice in areas where people are exposed to hot, moderate and cold environments and there has been recognition that air temperature is only one component of a human thermal environment.
The requirements of regional (e.g. European) and international (global) markets and the recognition that systems, services and products should be designed for human use have raised the profile of ergonomics and led to a proliferation of ISO standards, including those in the area of the ergonomics of the thermal environment. These include thermal comfort and are described below. For full details and practical applications the reader is referred to the original Standards.
ISO 7730 (2005) provides an analytical method for assessing moderate environments and is based on the predicted mean vote and predicted percentage of dissatisfied (PMV/PPD) index, and on criteria for local thermal discomfort. If the responses of individuals or specific groups are required, then subjective measures should be used (ISO 10551, 1995).
5.6.1 ISO 7730: Moderate Thermal Environments – Determination of the PMV and PPD indices and specification of the conditions for thermal comfort
This standard considers whole-body thermal sensation and local thermal discomfort caused by draughts. It is based on the predicted mean vote (PMV) and the predicted percentage of dissatisfied (PPD) indices (Fanger, 1970), and more recent work concerning draughts (Olesen, 1985; Fanger et al., 1989).
The PMV is the predicted mean vote of a large group of persons, on the following thermal sensation scale, if they had been exposed to the thermal conditions under assessment.
| + 3 | hot |
| + 2 | warm |
| + 1 | slightly warm |
| 0 | neutral |
| − 1 | slightly cool |
| − 2 | cool |
| − 3 | cold |
The PMV is calculated from the air temperature, mean radiant temperature, humidity and air velocity of the environment and estimates of metabolic rate and clothing insulation. It is derived from a heat balance equation for the human body combined with empirically determined equations which define sweat rates and mean skin temperatures which are within comfort limits. The equation is provided in Section 5.5.4 above. It is based on data from the exposure of 1300 subjects to various thermal environments.
The draught rating (DR) is expressed as the percentage of people to be bothered by draughts, where
where ta is local air temperature (°C), v is local mean air velocity (ms− 1), Tu is local turbulence intensity (%) defined as the ratio of the standard deviation of the local air velocity to the local mean air velocity.
The DR model is based on experiments on 150 human subjects for the following range of conditions:
It applies to people performing mainly sedentary activity with whole-body sensations close to neutral. Risk of draught is lower at higher activities and if people are warmer than neutral.
Thermal comfort is defined in the standard as ‘that condition of mind which expresses satisfaction with the thermal environment’. Dissatisfaction may be caused by whole-body or local discomfort. An annex (included, but labelled as not part of the standard) provides guidance in terms of levels of dissatisfaction. This includes dissatisfaction caused by whole-body discomfort and by draughts and other local effects (e.g. thermal gradients, asymmetric radiation, etc.). The Standard also considers the long-term thermal comfort performance of buildings and also provides guidance on appropriate conditions for different quality and spaces. Tables of metabolic rate and clothing insulation values are included. For more detailed estimates the user can use ISO 8996 (1990) and ISO 9920 (1992). A computer program is provided to allow ease of calculation and efficient use of the standard.
5.6.2 ISO 7726: Thermal Environment – Instruments and methods for measuring physical quantities
This Standard provides definitions of the basic parameters (air temperature, mean radiant temperature, humidity, air velocity) and derived parameters (natural wet bulb temperature, globe temperature). It also provides methods of measurement and specifications of measuring appliances.
No specific instrument is standardized, only specifications. The Standard can therefore serve as a guide to manufacturers of instruments as well as specifying measuring requirements, in a contract between investigator and a client.
5.6.3 ISO 10551: Assessing the influence of the thermal environment using subjective judgement scales
ISO 10551 (1995) presents the principles and methodology behind the construction and use of subjective scales, and provides examples of scales that can be used to assess thermal environments (see Table 5.7). A practical example and a discussion of methods of data analysis are provided. The principle of the Standard is to provide background information to allow ergonomists to construct and use subjective scales as part of the assessment of thermal environments.
Table 5.7
Subjective scales considered in ISO 10551 (1995)
| Judgement | Example | Related to |
| Perceptual | How do you feel now? (e.g. hot) | Personal |
| Affective | How do you find it? (e.g. comfortable) | Thermal |
| Thermal preference | How would you prefer to be? (e.g. warmer) | State |
| Personal acceptance | Is the environment acceptable/unacceptable? | Environment |
| Personal tolerance | Is the environment tolerable? |
5.7 Behavioural thermoregulation, thermal comfort and the adaptive model
People adapt to preserve comfort. When there is a ‘heat wave’ in the United Kingdom, enquiries are often received (from the press, public, etc.) on how to ‘keep cool’ in the hot conditions. Advice in the context of the six basic parameters could be to reduce clothing, increase air movement using fans, stay in the shade and be inactive. In extreme cases take a cool bath. It became clear, however, that there are simpler solutions based upon a paradigm shift in thermal comfort from that of traditional laboratory and field research. Cool places provide the opportunity to avoid becoming uncomfortably hot. People are not passive receptors of discomfort; if the opportunity is available to them they can take action including moving to more comfortable surroundings.
The most effective form of thermoregulation to ensure survival, comfort and performance is classically known as behavioural thermoregulation. Moving to more desirable thermal conditions, adjusting clothing, seeking shelter, opening windows, changing posture, cuddling, lighting of fires, switching on air conditioners or fans, and more are all examples of behavioural thermoregulation. The profound change in the six basic parameters due to behavioural thermoregulation demonstrates the potential of behavioural thermoregulation and emphasizes the continuous dynamic interaction between people and their thermal environments.
Although it has always been accepted that people are not passive, little account has been taken of human behaviour in design and assessment for thermal comfort. Attempts to influence accepted methods for the design and assessment of thermal comfort have used the term ‘adaptive modelling’ and proposals to use adaptive models of thermal comfort have been discussed for over forty years. The term ‘adaptive’ is unfortunate because it implies longer term, even evolutionary adaptation; however, it is now widely used in the context of thermal comfort and although requiring more stringent definition, it does not seem to cause confusion. Early ‘adaptive modellers’ include Auliciems (1981) from Australia and Humphreys and Nicol (1970) from the United Kingdom. The drive for adaptive modelling has continued with researchers from Australia, the United Kingdom, and the United States involving theoretical issues and worldwide field studies. The debate is on-going; however, the ASHRAE (1997) global database of thermal comfort field experiments and associated adaptive model (de Dear, 1998; de Dear and Brager, 1998, 2002) and the interest of the International Standards Organization (ISO) ensured that adaptive models are being given serious consideration. Clearly if we are to improve our understanding of ‘adaptive modelling’ there needs to be rigorous scientific investigation of this area based on the method of null hypothesis. However, the most important point about behavioural or adaptive modelling is the paradigm shift and the new opportunities that it affords to develop our knowledge of thermoregulation.
5.8 Equivalent clothing index (IEQUIV)
In any environment there will be opportunity to adapt to maintain thermal comfort and these adaptive adjustments (behaviours) can take many forms and can occur in combination (e.g. adjust clothing, change activity, open a window, change posture). Each of the actions and their combination will have an effect on human heat exchange. It is therefore possible to represent these effects in terms of the equivalent effect of changing one of the parameters in the heat balance equation. This could be any of the six basic parameters; however, a convenient approach would be to relate the total effect of all adaptive behaviour to the equivalent effect of adjusting clothing. An equivalent clothing index (IEQUIV) can be described as follows. The equivalent clothing index (IEQUIV) is the clothing insulation that would give equivalent thermal comfort to people with no adaptation as the thermal comfort of people who adapt to their thermal conditions.
A group of people initially wearing 1.0 Clo who change clothing, change activity, change posture and open a window to maintain a neutral thermal sensation may have an equivalent clothing index value of 0.2 Clo, where 0.2 Clo would represent the clothing insulation required to maintain a neutral thermal sensation in the original conditions. The total of all adaptive behaviour therefore summates to a reduction of 0.8 Clo. The equivalent clothing index value can then be substituted into the PMV equation, instead of the clothing insulation value, to give a PMV that takes account of adaptive behaviour.
Table 5.8 shows a possible relationship between adaptive opportunity and IEQUIV for a person dressed in 1.0 Clo.
where ISTART is the clothes worn before adaptation and IADJ is the change in clothing that would have equivalent effect on thermal comfort to the sum of all adaptive measures.
For cases where no adaptation is possible, then clothing, posture, activity, physical environment (e.g. windows) cannot be adjusted (IADJ = 0). Therefore the equivalent clothing is the actual clothing worn at the beginning of the exposure period (IEQUIV = ISTART). For maximum adaptation in the heat, it will be possible to take off all clothing as well as making other adaptive adjustments (open windows, reduce activity). For these conditions it may be useful to adjust parameters separately (e.g. metabolic rate) in the rational index (e.g. PMV) as well as using IEQUIV = 0. In these circumstances, thermal comfort is an unusual concept and, in practice, adaptive effects greater than taking off all clothing would be unusual. It may therefore be reasonable to assume IEQUIV = 0 is a minimum practical value. A similar argument applies for IEQUIV in cold conditions where doubling of clothing insulation is a realistic maximum with more detailed analysis required for special cases. It should be noted that IADJ may be different for hot conditions than for cold conditions as different adaptation may be required; for example, if opening windows is the only adaptive opportunity, then IADJ may be ‘medium’ in hot conditions but ‘minimum’ in the cold, etc.
5.9 Equivalent clothing index, the comfort temperature range and temperature limits in offices
Although a rational index such as the PMV provides a versatile tool with which to determine thermal comfort conditions in terms of the six basic parameters (variables), there is a requirement to provide temperature ranges within which people can maintain comfort. Suppose people in a building wear 1.0 Clo, perform sedentary activity in conditions with no radiant load (ta = tr), and vapour pressure of 1.0 kPa in still air. They can reduce their clothing but cannot increase it, they can move around and open windows. We could assess the adaptive opportunity as high in the heat and low in the cold. A PMV of 0 (neutral) is then obtained at 24 °C for IEQUIV = ISTART = 1.0 Clo; 28.5 °C for IEQUIV = 0.25 Clo and 22.8 °C for IEQUIV = 1.25 Clo. This provides a comfort temperature range of 22.8 °C to 28.5 °C. If a range between +1 (slightly warm) and –1 (slightly cool) is regarded as acceptable then an acceptable range would be 18.2 °C to 30.5 °C (see Table 5.6).
The PMV is an example of a rational index and its validity does not affect the principles of the above. The IEQUIV method is a behavioural adjustment which is a practical way forward. It is preferable to ‘expectancy’ adjustments where statements such as, ‘It is what they are used to and do not expect better’, border on the unethical and, in any case, are flawed as if people claim comfort at 30 °C, how do we know that they would not also claim comfort at PMV = 0 (e.g. 24 °C).
5.10 Sustainable thermal comfort
If we consider sustainability to be the use of resources in such a way that they do not become exhausted, or even reduced at all, this will provide a context for achieving sustainable thermal comfort. We have seen in this chapter that there are many ways of attaining thermal comfort. A useful starting point is to consider the many combinations of air temperature, radiant temperature, humidity, air velocity, clothing and activity that will create comfort conditions. Practical constraints and context will leave a subset of the conditions to provide thermal comfort in any design or application. It is reasonable to assume that a further subset of those conditions will provide sustainable thermal comfort.
Knowledge of what constitutes a sustainable ‘system’ or environment is in its infancy and enthusiasm for ‘saving the planet’ has not always been based upon ‘solid science’, clear rationale and evidence. By definition the degree of sustainability requires a view on resources used, resources available and the dynamics of the relationship between them. To specify conditions for sustainable thermal comfort therefore requires much more analysis and research than is currently available. There are, however, practical suggestions for ‘good housekeeping’ that can greatly reduce the use of resources and energy and decrease carbon production. Two obvious measures towards sustainable thermal comfort are to reduce use of resources due to technological control (e.g. air conditioning) and design environments that make it less resource intensive to create thermal comfort and complement, for example, human behaviour and adaptive opportunity to allow individuals to achieve thermal comfort.
The use of the IEQUIV index to take account of adaptive opportunities provides a rational approach to specifying thermal comfort. If the IEQUIV index is used in conjunction with the PMV thermal comfort model then a range of conditions can be found within which thermal comfort can be maintained. This simple approach can have dramatic effects in saving resources. For example, there is a tendency for people to over-react when faced with hot or cold conditions. In a heat wave, air temperatures are reduced to levels unnecessary for the achievement of thermal comfort when simple measures such as air movement or reduced clothing (e.g. taking off a tie) could be used. Indeed, such is the over-reaction that complaints of being too cold in rooms, due to air conditioning, are not uncommon when the outside weather is hot. There seems to be an over-reaction in this scenario which could significantly contribute to an increase in total cooling demand during a heat wave. In cold outdoor conditions it is not necessary to overheat a space when a simple increase in clothing can provide comfort. Parsons (2002) demonstrated that people could maintain comfort at below 20 °C air temperature using clothing, although local effects (particularly among females) suggested below 19 °C started to cause discomfort. The over-reliance on technology to provide thermal comfort solutions leads to high energy costs, due to excessive cooling with air conditioning or unnecessary use of heating; this can be considered poor housekeeping and represents an unsustainable approach to achieving thermal comfort.
There are clearly limits to adaptive behaviour for thermal comfort. However, a significant saving in energy and resources could be made and sustainable thermal comfort provided if we did not heat offices above 20 °C or cool them below 25 °C. This can be demonstrated by an IEQUIV and PMV analysis (see Table 5.6).
For more detailed values a full analysis, involving all six parameters, is required. However, cooling below 20 °C begins to cause local discomfort and above 25 °C an unacceptable reduction in clothing, or a tendency to become warm and sticky. Using simple measures, some major savings can already be made and help move towards sustainable thermal comfort. Further work is needed to our philosophical approach and technical and scientific understanding.
5.11 References
ASHRAEThermal comfort conditions. New York: ASHRAE, 1966. [standard 55.66].
ASHRAE, Physiological principles, comfort and health. Fundamentals Handbook, 1989. [Atlanta, USA].
ASHRAE, Thermal comfort. ASHRAE Handbook of Fundamentals, 1997. [Chapter 8, Atlanta, USA].
Auliciems, A. Towards a psychophysiological model of thermal perception. International Journal of Biometerology. 1981; 25:109–122.
Bedford, T., The warmth factor in comfort at work: a physiological study of heating and ventilation. Industrial Health Research Board Report No. 76, London, HMSO, 1936.
de Dear, R.J., A global database of thermal comfort field experiments. ASHRAE Transactions. 1998:104. [Part 1].
de Dear, R.J., Brager, G.S., Developing an adaptive model of thermal comfort and preference. ASHRAE Transactions. 1998. [Atlanta, USA].
de Dear, R.J., Brager, G.S. Thermal comfort in naturally ventilated buildings: revisions to ASHRAE Standard 55. Energy and Buildings. 2002; 34(6):549–572.
Drury, C.G. Methods for direct observation of performance. In: Wilson J.R., Corlett E.N., eds. Evaluation of Human Work – A Practical Ergonomics Methodology. London: Taylor & Francis; 1990:35–57.
Fanger, P.O.Thermal Comfort. Copenhagen: Danish Technical Press, 1970.
Fanger, P.O., Melikov, A.K., Hanzawa, H., Ring, J., Turbulence and draught. ASHRAE Journal. 1989. [April].
Guildford, J.P. Psychometric Methods, 2nd edn. New York: McGraw-Hill, 1954.
Hollies, N.R.S., Custer, A.G., Morin, C.J., Howard, M.E. A human perception analysis approach to clothing comfort. Textile Research Journal. 1979; 49:557–564.
Houghton, F.C., Yagloglou, C.P. Determining equal comfort lines. Journal of ASHVE. 1923; 29:165–176.
Humphreys, M.A. Clothing and thermal comfort of secondary school children in summertime. CIB Commission W45 symposium thermal comfort and moderate heat stress, Watford, London, HMSO, 1972.
Humphreys, M.A. Outdoor temperatures and comfort indoors. Building Research and Practice. 1978; 6(2):92–105.
Humphreys, M.A., Nicol, J.F. An investigation into the thermal comfort of office workers. Journal of the Institution of Heating and Ventilation Engineers. 1970; 30:181–189.
ISO 7726Thermal Environments – Instruments and methods for measuring physical quantities. Geneva: International Standards Organization, 1985.
ISO 7730Moderate Thermal Environments – Determination of the PMV and PPD indices and specification of the conditions for thermal comfort. Geneva: International Standards Organization, 2005.
ISO 8996Ergonomics of the Thermal Environment: Estimation of metabolic heat production. Geneva: International Standards Organization, 1990.
ISO 9920Estimation of the thermal characteristics of a clothing ensemble. Geneva: International Standards Organization, 1992.
ISO 10551Assessment of the influence of the thermal environment using subjective judgement scales. Geneva: International Standards Organization, 1995.
ISO DIS 28802Ergonomics of the Physical Environment – The assessment of environments by means of an environmental survey involving physical measurements of the environment and subjective responses of people. Geneva: International Standards Organization, 2009.
McIntyre, D.A. Indoor Climate. London: Applied Science; 1980.
McNall, P.E., Ryan, P.W., Rohles, F.W., Nevins, R.G., Springer, W.E., Metabolic rates at four activity levels and their relationship to thermal comfort. ASHRAE Transactions; 74, 1968. [Pt I. IV.3.1].
Moser, C., Kalton, G. Survey Methods in Social Investigation, 2nd edn. London: Heinemann, 1971.
Nevins, R.G., Rohles, F.H., Springer, W.E., Feyerherm, A.M. A temperature humidity chart for thermal comfort of seated persons. ASHRAE Transactions. 1966; 72(I):283–291.
Olesen, B.W., Local thermal discomfort. Bruel, Kjaer, eds. 1985. [Technical Review No. 1, Copenhagen].
Olesen, B.W., Sliwinska, E., Madsen, T.L., Fanger, P.O. Effects of body posture and activity on the thermal insulation of clothing: measurements by a movable thermal manikin. ASHRAE Transactions. 1982; 88(2):791–805.
Parsons, K.C. Safe surface temperatures. In: Lovesey E.J., ed. Contemporary Ergonomics. London: Taylor & Francis, 1993.
Parsons, K.C. The effects of gender, acclimation state, the opportunity to adjust clothing and physical disability on requirements for thermal comfort. Energy and Buildings. 2002; 34:593–599.
Parsons, K. Human Thermal Environments, 2nd edn. London: Taylor & Francis, 2003.
Rohles, F.H., Rohles, F.H., Thermal sensations of sedentary man in moderate temperature. 1970Institute for Environmental Research: Special Report. USA: Kansas State University, 1970.
Rohles, F.H., Nevins, R.G. The nature of thermal comfort for sedentary man. ASHRAE Transactions. 1971; 77(1):239–246.
Sinclair, M.A. Subjective assessment. In: Wilson J.R., Corlett E.N., eds. Evaluation of Human Work – a practical ergonomics methodology. London: Taylor & Francis; 1990:58–88.
Weston, J.C. Heating research in occupied houses. Journal of the Institute of Heating and Ventilating Engineers. 1951; 19:47–108.
Wyon, D.P., Halmberg, I., Systematic observation of classroom behaviour during moderate heat stress. Thermal Comfort and Moderate Heat Stress. Proceedings of CIB, W45 Symposium, Watford, London, HMSO, 1972.