5. Indoor climate robustness

5.1 Introduction

International field studies into the causes discomfort and building related heath symptoms in hundreds of office buildings show that occupants in certain building types are more satisfied than occupants in other building types. The studies show that the following risk factors lead to significantly more discomfort and health symptoms (Leyten and Kurvers, 2006):

Cooling of the supply air;

Humidification of the supply air;

Recirculation of room air11;

The absence of openable windows;

Insufficient possibilities for the occupants to adjust the temperature;

Working in large open plan offices.

Our own studies in office buildings in the Netherlands also show the following risk factors, which are closely linked to the above factors:

Controlling the indoor temperature within narrow limits;

No openable windows or windows that, although openable, are not practically usable by the occupants at their own choice;

Insufficient thermal mass to store excessive heat;

Large areas of glass in the facade trap too much solar heat;

Lack of controllable external sun shading;

The last two often occur in combination.

Various causes for the poor performance of buildings are mentioned in the literature. Examples are that more complex air handling systems contain more potential sources of chemical and microbiological indoor air contaminants in filtering, cooling and humidifying sections (Weschler, 2004; Clausen et al., 2011). Furthermore, it is reported that more complex systems are more susceptible to failure due to inadequate maintenance (defective parts of installations are often discovered only after a long time), inadequate commissioning of the building systems, altered use compared to the design, or cutbacks during the construction phase that cause the resulting situation to no longer function as intended in the design.

In buildings where the occupants perceive the indoor climate as uncomfortable, the energy consumption is often higher than anticipated during the design process (Turner, 2008; Van den Ham, 2009). These are often buildings with (complex) HVAC systems. Buildings are also being realized that aim to meet a high sustainability ambition and strive for an energy-efficient design, but where the emphasis is on innovative technical installations and less on building physics, architectural and passive design. What many of these designs have in common is that little consideration is given to the outdoor climate, for example through limited thermal mass and glass façades with a high level of solar radiation and no external sun blinds.

5.2 The concept of robustness

In order to explain the relationship between building characteristics and the discomfort and health symptoms of occupants, this chapter describes the concept of indoor climate robustness (Leyten and Kurvers, 2006, 2011). A robust system is a system where the output is relatively insensitive to variations or uncertainties of the input. The term robustness is used in computer science, statistics and economics, among others. Robustness should not be confused with resilience, which is “the ability to bounce back after disruption”. Unlike robustness, which is proactive, resilience is reactive, following incidents in which system performance has already been affected. 

The robustness of the indoor climate of an office building, including the architectural properties and its air handling systems, is defined as the degree to which the realised indoor climate of the building meets its design goals regarding comfort, health and energy consumption during the operational phase. Buildings differ in their robustness: some meet their design goals better than others. Designs that appear to perform flawlessly at the design stage in simulation calculations and test chamber studies may systematically perform less well in practice. This is part of the explanation for the large differences in occupant satisfaction between different types of buildings mentioned in Section 1.1. Robust climate design indicates that the indoor climate and energy use in daily use meet the design objective: an indoor climate that is as comfortable and healthy as possible, with the lowest and most predictable energy use. In the following sections, a number of mechanisms are mentioned that can increase the robustness of the building and its systems, or, in other words, reduce the risks of comfort and health complaints and high energy consumption. It will also become clear that some mechanisms overlap to some extent.

5.3 Robustness and technical functioning of buildings

There is a tendency that buildings and climate systems become more and more complex, because of increasing demands on the quality of thermal comfort, reducing energy and the application of renewable energy. The quality of thermal comfort is often based on incorrect assumptions what thermal comfort actually is and how to achieve this (see Chapter 7). The integration of ICT in the built environment increases the possibilities for detailed control and gathering of information about the building, but also make it more complex and difficult to understand it during the design and operational phase. But also for the occupants if the operation is not straightforward.

Passive instead of active solutions

A strategy for achieving both a good indoor environment and low energy use is recommended by Roulet (2006): control the indoor environment as far as possible by passive means, and use active means only to adjust the indoor environment if necessary. Examples of this strategy are controlling the sources of indoor air pollutants instead of using increased ventilation and controlling the temperature by managing heat sources and applying passive building physics principles instead of opting for mechanical cooling. These two examples are further explained below.

Control of sources of indoor air pollution

The control of pollution sources has priority over increased ventilation or personal protection (CEN, 1998). From a purely theoretical point of view, this is not obvious. If the objective is to limit exposure to a certain maximum, one might assume that this can be achieved just as well by increased ventilation or personal protection as by source control. But from a robustness point of view, there are good reasons to prefer source control. The most obvious reason is that if the source is removed one can be very certain that no exposure is taking place. This is in contrast to increased ventilation where exposure remains possible in practice due to problems such as underestimation of source strength, inadequate ventilation rates or inadequate ventilation effectiveness.

Control of heat sources by passive building physics principles

The strategy of using as many passive means as possible to control the indoor environment and using active means only to fine-tune the indoor environment where necessary also applies to indoor temperature control. In parts of the world, this strategy is used for winter temperature control. The standard is to increase the thermal insulation of the façade as much as possible to minimise the need for active heating. Following the same strategy in summer would mean minimising the solar heat load and internal heat sources and maximising the use of thermally effective building mass to eliminate or reduce the need for active cooling. This strategy has not been widely accepted; on the contrary, it has become common practice to accept high solar heat loads by omitting external shading, usually due to cost considerations or architectural reasons. In addition, a low thermally effective mass is often chosen to allow for flexible layout. All this leads to a higher cooling demand, requiring more extensive HVAC systems.

Insensitivity to deviations from design assumptions

Certain building designs are sensitive to (minor) deviations from design assumptions. An example is induction units, where it is very important that the properties of the unit are accurately tuned to the properties of the room. An incorrect adjustment can lead to distortion of the flow pattern and thus, for example, to too high air velocities in the occupied zone. It is our experience that such mismatches occur regularly in practice. If, during the construction process, a different air supply grille is chosen or a different ceiling structure is used, the flow pattern may be disturbed. Technically simpler heating and ventilation systems are less sensitive to such changes and therefore more robust12.

Feasible maintenance requirements

Some designs require more maintenance than others. To contrast two extremes: a larger heat-accumulating building mass, for example by using thermally open ceilings, to limit temperature excursions through passive cooling, is a robust measure. Once installed, it requires hardly any maintenance. Another robust choice is to reduce the window area to limit solar heat gain. On the other hand, the cooling section and variable volume boxes, for example, require periodic maintenance. Particularly high maintenance requirements are imposed by devices where moisture can lead to bacterial growth, such as spray humidifiers (ASHRAE Handbook, 2016). Such devices reduce the robustness of the HVAC system. The robustness can be further reduced by components such as the condensation drains of induction units, which can be contaminated with bacteria (Byrd, 1996; Menzies et al., 2003; Asikanen et al., 2006). Such facilities are scattered throughout the building, which significantly reduces the likelihood that they will all be properly maintained. In general, passive solutions require less maintenance and are therefore more robust than more complex mechanical systems, and centralised systems are more robust than decentralised ones.

Separation of heating and ventilation

If heating and ventilation are integrated in some way, they seem to be more prone to malfunction than systems where heating and ventilation are separated as much as possible. For example, with induction units, reducing the air supply to avoid drafts or noise can also reduce the heating or cooling capacity. Another example is variable volume systems, where controlling the air supply for the purpose of keeping the indoor temperature constant can lead to insufficient fresh air supply and therefore inadequate air quality.

No time-varying flow rate

One of the conclusions of the European IAQ Audit (Bluyssen, 1995) was that in systems with recirculated exhaust air, the actual amount of recirculated air was often higher or lower than the specified amount. This poses the risk that with recirculation, the fresh air supply is actually lower than intended. This not only highlights the risks of recirculation, but also reminds us that, in practice, supply air volumes are not controlled as precisely as we like to think. In other words, controlling supply air volumes reduces robustness.

A system that can function reliably in practice in, for example, schools and homes is CO2-controlled ventilation. The air is supplied via (pressure-controlled, draught-free) grilles in the facade and discharged centrally via a CO2-controlled mechanical ventilation unit. It is of course essential that the CO2 sensor functions correctly.

5.4 Robustness and interaction between occupant and building

Transparency and control for occupants

An air handling system is transparent when the occupants can acquire a certain basic understanding of how the system works just by looking at it and using it, and when they can see for themselves when the system is not working properly and to some extent know what is wrong with it. Examples of this are:

Most people have a basic understanding of how heating radiators work. Malfunctions of radiators can be noticed by occupants: the radiator does not warm up, the thermostat knob cannot be turned, or it does not seem to affect the temperature;

The operation of an openable window is also understandable for most people. One can immediately see how far it is open, as well as whether or not different positions can be set. If opening windows cause draughts, it is usually easy to identify which windows are affected;

If an exterior sunshade is not working properly, occupants will notice this immediately and building management can take action.

With complex climate systems, it is sometimes difficult even for experts to get an idea of what is causing occupants' discomfort.

Occupant control

Transparent and effective user control of the indoor environment increases robustness for two reasons:

It allows occupants to adapt the indoor environment to their own preferences and to variations over time;

Within certain limits, it allows the occupant to compensate for any deficiencies in the functioning of the building and the HVAC system.

Balancing positive and negative impacts through occupant control

Control options are especially effective for users if they can balance the various positive and negative consequences of their interventions (Clausen & Wyon, 2006). An example of this is described in Section 4.5 where it was found that in naturally ventilated buildings, when using windows to open, users can always make a trade-off between the need for cooling by airflow and fresh air on the one hand and reducing noise from outside on the other because in these buildings both thermal comfort and noise levels are bonus factors.

Natural user behaviour leads to improvement of the indoor environment

If users can control their environment, for example by regulating the temperature or opening windows, and if the number of workplaces per room is not too large, the natural behaviour of users will lead to an optimal situation for most users. An example of a situation where the natural behaviour of the users leads to a worse situation for themselves and/or for others is when there are draught problems due to the mechanical air supply. In some cases, users therefore tape over ventilation grilles, reducing the ventilation from that particular grille, with a risk of inadequate air quality, and increasing the air velocities at other grilles, which can lead to more draught complaints there.

The environmental Gestalt promotes acceptance

The environmental Gestalt13 encompasses the entire context of the indoor environment and the user experiences in these areas (de Dear, 2004). Formulated in this way, this may seem too abstract, but it can be defined in terms of transparency and control of the indoor environment and the concept of fairness. An environmental Gestalt promotes acceptance if the following conditions are simultaneously met (Leyten, Kurvers & van den Eijnde, 2009):

Deviations from a comfortable situation can be feasibly reduced or compensated for by occupants with limited negative side effects for themselves or others. This includes the ability to influence the environment and choices about personal factors such as activity and clothing. If there are negative side effects, the possibilities for influence should enable occupants to weigh up the positive and negative consequences of choices;

The remaining deviations from a comfortable situation are then understandable to the occupant through the transparent functioning of the building and its systems;

The remaining deviations from a comfortable situation are considered fair by the occupants on the basis of insight through transparency and a feeling of co-responsibility for the working environment that arises from the occupants' ability to influence the situation.

Additional conditions that promote acceptance

A lack of privacy and views can contribute to dissatisfaction. Good privacy and good views improve user well-being and reduce stress. Workers with more routine work need at least as much privacy as those with creative work because they already have little influence on the work they do. A good view includes a visible skyline and the ability to see distant objects, vegetation and weather conditions (Vroon et al., 1990). That taking complaints seriously by management increases acceptance is something we have seen in complaint assessment surveys we have conducted and follows from the social psychological fairness theory (Whitley et al., 1995). Privacy and views are determined by the building design and the resulting possibilities for arranging the workspace layout.

5.5 The individual office room as a robust and accepting environmental gestalt

Since the 1960s, the layout of offices has changed from smaller cellular office rooms with 1 to 4 workplaces to larger spaces, which are usually referred to as open-plan or landscape offices. In recent years, new office concepts have been introduced such as flex spaces, mixed offices, combination offices or eco landscape offices. The aim is a flexible office design with additional claimed benefits such as better communication, better social interaction and more responsibility and freedom for employees. In addition to the advantages of open-plan offices, such as saving space through a greater density of people and more flexible (re)arrangement, also better cooperation, communication and a related improvement in productivity are suggested. However, there is no scientific basis for this. On the contrary, research by various disciplines, such as architecture, engineering and psychology, show negative effects of open-plan offices on the perception of the office environment (Vroon et al., 1990; Pejtersen et al., 2006). The negative effects are felt in various aspects, such as general dissatisfaction with the office environment, decreased concentration and loss of privacy (Sundstrom et al., 1982; Leaman & Bordass, 1999; Kaarlela-Tuomaala et al., 2009). Many researchers consider open plan offices to be one of the main underlying causes of symptoms of sick building syndrome, such as headaches, fatigue, difficulty concentrating, eye irritation and respiratory complaints (Klitzman & Stelman, 2006; Pejtersen et al., 2006; Wittersey et al., 2004). Noise nuissance causes concentration difficulties, dissatisfaction and reduced privacy (Danielsson & Bodin, 2009; de Croon et al., 2005).

The development of large databases containing results of occupant surveys is making it possible to carry out increasingly reliable analyses. In order to investigate whether the claimed advantages of open plan offices (better communication and cooperation) also exist in practice and outweigh the reported disadvantages (concentration problems, noise disturbances and health symptoms), the results of 303 offices with 42,764 responses were analysed (Kim & De Dear, 2013). This study shows that none of the claimed advantages of open plan offices are based on fact. On the contrary, satisfaction with “interaction between office occupants” were found to be higher in enclosed private offices. The negative effects of noise nuisance and lack of privacy outweighed the claimed “ease of interaction” in open-plan offices. Satisfaction with the office environment was highest in enclosed private offices.

More field studies show that physical symptoms and dissatisfaction with the indoor environment occur more frequently as the number of workstations per workspace increases (Leyten & Kurvers, 2007). These results are supported by Pejtersen et al. (2006), who show that the more workstations per workroom, the greater the number of complaints about thermal comfort, indoor air quality, noise, fatigue, headaches and concentration difficulties occur (see Figure 5.1).

Figure 5.1: Correlation between the number of workstations per workspace and dissatisfaction with the thermal environment (top left), dissatisfaction with the air quality (top right), dissatisfaction with the acoustic environment (bottom left) and concentration problems, fatigue and headaches (bottom right). Source: Pejtersen et al., 2006.

This can be explained within the building robustness hypothesis: in the enclosed private offices, most of the above-mentioned robustness mechanisms apply, and they all work in the right direction:

Due to its limited dimensions, especially its limited depth, and its high surface/volume ratio, enclosed private offices can be equipped with a less complex heating and ventilation system with higher robustness;

If every room is equipped with windows that can be opened and temperature control for the winter period, this will ensure maximum transparency and control by the occupants. An illustration of this is a study in 8 office buildings in The Netherlands (Kurvers, Van der Linden & Boerstra, 2002) which shows that the perceived ability to open windows decreases as the size of the office spaces increases. The percentage of people who can open the window as needed decreases from 79% in single rooms to 27% in rooms with more than four people (Figure 5.2);

Figure 5.2: Experienced ability and usability to open windows in relation to group size, in 8 office buildings in The Netherlands. Source: Kurvers, Van der Linden & Boerstra, 2002.

The presence of openable windows instead of mechanical cooling will increase the acceptability of higher temperatures in summer and make the application of adaptive thermal comfort standards more feasible. It also makes thermal comfort a bonus factor (See Section 4.5)

A small office room encourages users who share a space to be considerate of each other, for example when making telephone calls. This also includes the acceptability of temporary disturbance as the reason and necessity are understood;

The small office space has a positive effect on social relationships in general. This has been demonstrated by research into the social effects of open plan offices in the 1960s and 1970s (Vroon et al., 1990; Pejtersen et al., 2006);

Compared to open plan offices, small room offices reduce noise nuisance from the workspace, give more privacy to the users and generally provide a better view of the outside, as the desk is usually closer to the window. All of this will enhance the acceptability of the working environment as a whole.