Outlook for the ISO 22955: Acoustic quality of open plan offices Thomas Bonzom 1 CARSAT LR 29 cours Gambetta, 34068 Montpellier cedex 2, France Patrick Chevret 2 INRS 1, Rue du Morvan, CS 60027, 54500 Vandœuvre-lès-Nancy, France Laurent Brocolini 3 INRS 1, Rue du Morvan, CS 60027, 54500 Vandœuvre-lès-Nancy, France Yoan Le Muet 4 SAINT-GOBAIN ECOPHON AB Yttervägen 1, 265 75 Hyllinge, Sweden

ABSTRACT The ISO 22955 standard « Acoustic quality of open plan offices » was published in may 2021. It aims to provide principles, descriptors, and measurement methods to characterize acoustics, which are easy to use and correspond to the perception of the acoustical environment by the occupants of the spaces. The scope of the standard is the acoustics of open-plan offices and, more specifically, cognitive effects of noise, i.e. acoustic comfort and noise disturbance in association with the activity. After more than a year of use of the standard, this document proposes to take stock of it and to identify potential paths of reflection for a possible future revision. Among the points of discussion considered, are the consequences of the evolution of office work, activities and concepts in the tertiary sector (flex office, co-working, teleworking, etc.), and their implication on acoustic quality. Another fundamental matter is the relevance of the indicators used in the standard and the associated required values.

1 thomas.bonzom@carsat-lr.fr 2 patrick.chevret@inrs.fr 3 1aurent.brocolini@inrs.fr 4 yoan.le-muet@saint-gobain.com

Jai. inter noise 21-24 AUGUST SCOTTISH EVENT CAMPUS ? O ? . GLASGOW

1. INTRODUCTION

The ISO 22955 standard “Acoustics — Acoustic quality of open office spaces” [1]. was published in 2021. It was expected for a while by the office acoustics community, and it seems that it was overall very well received. The standard puts a focus on understanding the activity as a preliminary to the acoustic design. To accompany that approach new ways to set up objectives and a new acoustic indicator were developed and used within the standard, specifically useful for open office spaces combining multiple activities. As this is the first version of this standard, and because this was a relatively new approach, there might be room for improvement, especially now that acousticians around the globe have started using the standard and are coming up with feedback from the field. Concurrently, the reality of office work quickly evolved, pushed by the pandemic, and open plan offices are seeing new challenges which might need to be accounted for.

2. PRINCIPLES OF THE ISO 22955 STANDARD

The objective of the ISO 22955 standard is to bring open offices users, a good acoustic environment, by focusing on different types of office work activities and the acoustic conditions they require. To do so it provides a complete technical and methodological guidance along with practical tools such as layout and room acoustic guidance, surveys, process flowcharts, to help stakeholders achieve this goal. It is intended to support design and planning decisions from conception to use being on a new project or refitting projects of existing business sites . It aims also to provide basis for communication between all the stakeholders involved in the process: end customers, project owners, prescribers, consultants, etc.

2.1. A standard based on the work activity

The ISO 22955 standard puts the emphasis on the necessity to know and understand the work activity before setting acoustic criterions. Only then it proposes acoustic indicators and values, for each type of identified activity, to tackle eventual issues.

The type of activities identified in the standard are given below: • Space type 1: activity not known yet – vacant floor plate;

• Space type 2: activity mainly focusing on outside of the room communication (by telephone/audio/video); • Space type 3: activity mainly based on collaboration between people at the nearest workstations;

• Space type 4: activity based on a small amount of collaborative work; • Space type 5: activity that can involve receiving public; • Space type 6: combining activities within the same space.

Space type 1 was introduced to allow the inclusion of basic acoustic challenges in early stages of construction of office buildings by property developers or a landholding trust. At this stage the activity is generally not know.

2.2. Different acoustic challenges from workstation to floorplate

The standard also reminds us that the acoustic challenges are different depending on the perspective. For instance, if the activity is mostly based around oral communication through telephone or video conference the challenge at the workstation is to have good intelligibility of the interlocutor. Between workstations the challenge becomes to not be disturbed (or generate disturbance) by intelligible conversations of neighboring workstations. On the floorplate the challenge is to keep the control of ambient noise which is a source of stress and fatigue for office workers. To address this the standard gives acoustic indicators and values adapted to the scale for each type of activity. The main indicators used are:

• at the workstation: noise level L Aeq,T

• between workstations: in situ attenuation of speech D A,S • on the floorplate: reverberation time T r , spatial decay rate of speech D 2,S , sound pressure level of speech at a distance of 4 m L p,A,S,4 m

2.3. Combining multiple activities, a new approach

Because a lot of open offices combine nowadays multiple activities, the concept of activity-based working being the culmination, the standard also proposes to tackle the risk of disturbance between different activities, which obviously need to be well identified prior to beginning this process.

As per a more classical approach when combining activities within a same open space, it is fundamental to question the acoustic challenges to each activity and to understand which activity is sensitive to noise and which activity is at greater risk of generating undesirable noise.

Based on the noise sensitivity of the receiver and on the liveliness of the source, a new approach [2] was developed by some members of the working group in charge of drafting the standard. Their work allowed the standard to propose a table values of in situ acoustic attenuation of speech (D A,S ) for various source/receiver combinations.

3. DISCUSSION EXISTING ITEMS OF THE STANDARD

3.1. D A,S measurement and target values

3.1.1 Definition

According to the standard, D A,S is the difference, in decibels, between the A-weighted spectrum of the normalized speech source (according to ISO 3382-3 revised in 2021) and the A-weighted sound pressure level at the receiving point.

D A,S = L A,1m,S – L A,n,S (1)

This indicator is also used to characterize the attenuation between two areas of the same open plan office in which different activities may take place. Figure 1 illustrates the two modes of use of the D A,S attenuation indicator. In the first mode (blue), the D A,S is used to assess the attenuation between two workstations in the same area. Here, the value of 6 dB(A) is the value recommended by the standard for a low collaborative activity. In the second mode (in red), it is used to evaluate the attenuation between two different activity areas. The value to be achieved depends on the activities carried out in the two zones. Here, the value of 18 dB(A) is the value recommended by the standard for a "collaborative" emission zone and a "low collaborative" reception zone. Attenuation targets can be achieved by spacing workstations sufficiently far apart, by installing ceilings of very good acoustic quality (class A) and by placing high and sufficiently thick partitions or screens between workstations.

Figure 1: Illustration of the two modes of application of the D A,S . Mode 1: between two neighboring workstations within the same activity zone (blue). Mode 2: between two different activity zones (red )

3.1.2 Measurement procedure

The D A,S is measured in two steps. First, the source is calibrated in an anechoic room and then the field work is carried out. The measurement is carried out with an omnidirectional source called reference source (R).

1st operation, before the measurement in the field

This is a free-field calibration operation, carried out in an anechoic chamber for example, during which the pressure level in each octave band i is measured at 1 m from the reference source (R): L p,1m,R,i .

2nd operation, in the field

The reference source (R) is placed at the workstation corresponding to the speaker's position (1.2 m from the ground) with the same orientation in relation to the microphone as in the anechoic chamber. The measurement microphone is placed at the listener's position (1.2 m from the ground). The pressure levels are measured per octave band with the same source settings as in the anechoic chamber: L p,n,R,i .

Once the measurement has been made, the difference for each octave between the sound pressure level measured in an anechoic chamber and the sound pressure level measured in the field is calculated. This difference corresponds to the sound field attenuation for the reference source:

D n,R,i = L p,1m,R,i – L p,n,R,I (2)

Then, the sound level that would have occurred at the listener's position with the normalized speech source is estimated by subtracting D n,R,i from the level at 1 m of the normalized speech source:

L p,n,S,i = L p,1m,S,i – D n,R,I (3)

The L p,1m,S,i values are provided in the ISO 3382-3 (2021) standard. The overall A-weighted level is then calculated for the speech source at 1 m and for the speech source level at the listener's station:

଻

௅ ೛,భ೘,ೄ,೔ ା஺ ೔

𝐿 ஺,ଵ௠,ௌ = 10 × log ൭෍10

൱ (4)

ଵ଴

௜ୀଵ

And,

଻

௅ ೛,೙,ೄ,೔ ା஺ ೔

൱ (5)

𝐿 ஺,௡,ௌ = 10 × log ൭෍10

ଵ଴

௜ୀଵ

Finally, the weighted attenuation A of the normalized speech source is calculated using the relationship (1).

3.1.3 Target values

The D A,S target values recommended in the standard are based on Harvie-Clark's [2] application of the Liveliness principle defined by Vellenga et al [3]. They assume that the annoyance experienced is related to the liveliness of the noise environment, which in turn is dependent on the ambient noise level ( L Aeq ) and the temporal variations of the latter ( L 5 - L Aeq ). To estimate a relationship between these two objective quantities that can be measured in the field, and the Liveliness , the authors proposed a scale ranging from Quiet to Turbulent and subjected more than 100 participants to an online evaluation of 12 office noise environments. The result is presented in Table 1, which gives the score for liveliness as a function of the pair ( L Aeq , L 5 ).

unug bay, (wiep 7

Table 1: Liveliness as a function of L Aeq and L 5 -L Aeq (over 5 min of recording). Figure taken from Vellenga et al. [3]

Although the values of the different parameters of the method were not provided by the authors, they do indicate an increment of 3 dB between neighbouring L Aeq classes (x2 = x1 + 3 dB, x3 = x2 + 3 dB, etc.) This table seems to show that there is a direct link between Liveliness and the objective parameters of the ambient noise.

Harvie Clark, on the other hand, assumes target values of Liveliness for the different types of activity proposed in the standard. These values are given in Table 2.

gy, < ¥,

Activity type Liveliness target value

Social and welfare 8

Informal meetings 7

Outside of the room communication (phone) 6

Collaborative 6

Non-collaborative 5

Focused phone 4

Focused individual work 3

Table 2: Liveliness target values for the different activities defined in ISO 22955 (table taken from Harvie-Clark et al. [2])

From the target values given in Table 2, Harvie-Clark calculates a target value for the signal-to-noise ratio in the area of the disturbed activity between the noise level in that area and the level from the neighboring area. For this purpose, he sets the following relation:

SNR target = 3 x (Liveliness target (disturbed activity) – Liveliness target (disturbing activity) ) (6)

However, the authors do not provide information on how this relationship was established. From the selected target values, it is then possible to deduce the target values for D A,S based on the relationship:

D A,S target = L A,1m,S – SNR target –L b target (7)

Where L b target is the target ambient noise in the disturbed area. The values of L b target are provided in the standard.

In the end, the target values recommended by the standard are given in Table 3.

Disturbed activity

D A,S (dB)

Outside of the

Focused individual

Informal meetings

room communication

Collaborative Non- collaborative

Focused

phone

work

(phone)

Social and welfare 15 15 18 24 27 32

Informal meetings 15 12 15 21 24 29

Outside of the room communication (phone)

12 18 21 29

Disturbing activity

Collaborative 18 21 26

Non- collaborative 18 23

Focused phone 21 23

Table 3: Recommended target values for D A,S (taken from [1])

Even if this approach of evaluating the annoyance produced by an activity on a neighboring activity using an objective indicator based on attenuation between zones is in itself particularly interesting, the target values recommended have yet to be consolidated. This is for three main reasons. The first one is that the Liveliness scale has not been shown to be related to the discomfort or fatigue felt by people, and it is the psychological factors that interest us in the end. The second reason is related to the lack of information concerning the relationship between the SNR target and the Liveliness target of the transmitting and receiving areas (Eq. 6). The third reason is also related to the lack of information about the target values of the Liveliness for each zone.

It is therefore useful today to consider a potential update of these values. As an example, a different approach to setting SNR target is proposed by Lenne et al [4]. This approach is based on direct laboratory assessment of cognitive annoyance and fatigue. This approach has already been used several times by the authors to study the irrelevant effects of speech noise in open plan offices. In this study, the authors investigated the significant effects on different psychological factors of noise associated with disruptive activities and SNR in the reception area. Based on the results, an update of Table 3 is proposed.

3.2. Use of artificial masking noise

The standard addresses the issues of noise comfort, in particular through the concepts of discretion and distraction reduction. The approach chosen is to limit the disturbance between adjacent workstations but also to optimize comfort for short-distance conversations. This document provides an opportunity to reflect further, by including an analysis of activities that involve more or less collaboration on the one hand, and by addressing everything that constitutes an open-plan space on the other, in particular in terms of surface treatments and additional office layout such as furniture, acoustic screens or low dividers, etc. but without the use of sound masking. Several reasons were given by the experts of the standardization group to exclude from the standard the situations for which a masking system would be used. The main reason is that, even if it is accepted that sound masking makes it possible to reduce the intelligibility of conversations and the distraction produced by them, the scientific work on the subject did not make it possible at the time of drafting the standard to reach a consensus on the real effectiveness of sound masking in real situations. It was therefore established that further studies needed to be carried out. The second reason is that the technologies used to create artificial background noise should be guided by specific guides or standards that would define the recommendations for installation, the maximum levels to be achieved, the distribution of loudspeakers, the type of sound signal that can be used (white noise, pink noise, noise imitating nature, etc.), the possible control of the background noise level, etc.

Since the standard has been issued, several studies have looked at the effects of sound masking on annoyance, performance, concentration, etc. In addition, a special session was held at the Internoise 2021 conference in Washington. Key publications in scientific journals include the work of Zang et al [5], who studied the effects of different types of masking signals for different SNRs in the laboratory. Hui et al. [6] studied the performance of a speech masking system where the masker design adapts to the spatial characteristics of target speech in order to create a more effective masker, while maintaining annoyance at a reasonable level. Zarei et al. [7] studied the influence of the homogeneity of the masking field in the room. Other works focuses on the soundscape approach which has attracted the attention of architects and urban planners [8]. Although this approach deserves special attention, it is outside the scope of the standard, which aims to control the distance of distraction, as soundscaping has more of an emotional psychological role in making the sound environment more pleasant. This list of publications is not comprehensive, and it is to be expected that others will come out before the next revision of the standard. The bibliographical work carried out by the standardization group during the drafting of the standard must therefore be continued in order to identify new results and possibly make the necessary changes. Particular attention should be paid to the long-term effects, particularly cognitive fatigue, associated with the use of masking noise.

4. EVOLUTION OF OFFICE WORK, THE NEED FOR NEW ITEMS?

As mentioned previously, the ISO 22955 standard defines several types of open offices according to the activity performed therein. If the standard makes recommendations to set good acoustic quality of the offices for these activity typologies, it does not however take into account the work organization (spatial and temporal). Yet, today, there are many ways to organize the office workspace, whether spatially or temporally. Indeed, new ways of organization have appeared during the last years: activity-based office (already in the standard), flex-office, teleworking or remote working, co- working, etc.

It is not unusual today for many companies to associate intellectual workers (or more generally office workers) with flexible forms of work, freed from space and time constraints. The worker must perform different tasks during the day (requiring concentration, meetings, business meetings, etc.) and must occupy different spaces according to his needs and the available resources. He is autonomous, flexible, responsible, etc. The emphasis is now on research and experimentation of new activity models based on concepts such as flexibility, reactivity or adaptability of social, technical and human resources [9].

Moreover, this search of flexibility and modularity has been encouraged by successive crises, both economic (the subprime crisis in 2008) and especially the latest pandemic (Covid19 2020-2022). Although it is difficult to predict the future of open offices, this crisis has forced companies to quickly reorganize their work models (generalization of teleworking and virtual teams, development of nomadism, workspace management, deployment of organizational agility, etc.) [9].

Today, more and more organizations are choosing to deploy flexible office solutions, i.e., shared offices, unassigned workstations, and open spaces.

The vocabulary used in the literature to designate and qualify these different solutions is very diverse: "flex-office", "non-territorial space", "dynamic work environments or offices", "activity-based", "self-service", "semi-partitioned spaces". Other examples and older terminologies may be found in the literature. Terms like "hot-desking", "just-in-time office" are also commonly used.

Although extremely diverse, the idea behind this vocabulary is always the same, namely a conciliation of the notions of work activities and spaces. The latter is presented as a shared resource that is "ready to use" and that professionals mobilize freely, flexibly, according to their needs. The different names used in the literature are difficult to define and do not allow us to establish a real typology [10]. They represent a great diversity of spaces according to the organizational ways set up within a wide variety of companies. However, it appears that many companies are turning to these new organizational models and are in search of more and more flexibility.

The sharing or non-allocation of individual workstations is one of the principal functional characteristics of the "flex office". This organizational way is presented as cutting-edge by the companies that use it. It is a device that allows for a combination of teleworking and "self-service" offices, facilitating the establishment of a rotation of personnel and optimizing the occupation of space. The benefits to employees are that they can partially work from home and see a reduction in the time spent in transportations.

Moreover, the shaping of the workspace seems to have a new role: the distribution of work activity in multiple space-time (the "anywhere", "anytime" work). These new flexible workspaces thus organize the rupture between space and work [9].

Through this question of the flexibility of these dynamic work environments, the place of space as a support for intervention on work and work activities may seem complex. However, it opens the necessary reflections for the implementation and co-construction of new forms of organization. The research for new ways of thinking about space is present in many companies and it seems necessary to integrate this notion in the standardization, especially so that these new organizations are implanted while considering the acoustic quality.

Regardless of how work is organized in open offices, one issue remains critical in all spaces and should be more explored further in the future of the standard. This is the area per occupant. Indeed, acoustics, in particular the level of ambient noise and the level of speech intelligibility, as well as the perception of sound environment are directly linked to the density and occupancy area of the space. The constant search for flexibility and job rotation, especially through teleworking, could lead companies to think anew their needs in terms of real estate and reduce the size of their offices. This means that the area per occupant could remain the same as today, or even decrease. So, this issue of area per occupant is becoming more critical and should be put forward in the standard.

Another topic related to workspace organization is co-working spaces. The co-working spaces are a type of space for which a very particular attention could be paid. Indeed, this type of workspace is very often a room where one or several individuals or companies, of different nature, activities and functions, rent spaces/furniture for more or less long periods. This type of space tends to multiply but is often not acoustically designed to accommodate multiple activities. Following the current trend, they are presented as favouring knowledge sharing, conviviality, and collaboration between professionals. However, these spaces are not always designed in an optimal way from an acoustic point of view. It would therefore be appropriate to take into consideration this type of working space and to issue recommendations for it.

5. CONCLUSIONS

ISO 22955 is considered by some to be groundbreaking as it brings new and old concepts together and shed lights on the importance of considering the working activity. It is certain that it filled a void as a standard covering this widely open offices acoustics had been long awaited.

For these reasons and also because office work is constantly changing, with new office design concepts making way and with recent events shaking up the reality of office workers and companies alike; there might be a need to consider a revision of the standard.

Obviously now is a little early to make such a move as the standard was just published last year, but it is never too early to start discussing and compiling ideas on how to improve an already great standard!

6. ACKNOWLEDGEMENTS

The authors would like to acknowledge the contributions and the great work that was done by the members of ISO/TC43/SC1/WG65 to produce a great standard and hope that they can continue working together to make it even better in the future.

7. REFERENCES

1. ISO-22955, Acoustics — Acoustic quality of open office spaces. 2021. 2. Harvie-Clark, J., Larrieu, F., & Opsanger, C. ISO 3382-3: Necessary but not sufficient. A new

approach to acoustic design for Activity-Based-Working offices. Universitätsbibliothek der RWTH Aachen, 2019. 3. Vellenga S, Bouwhuis T, Höngens T. Proposed method for measuring “liveliness” in open-plan

offices. 2017:9. 4. Lenne L., Chevret P., Parizet E. Assessment of annoyance and cognitive fatigue of open-plan

office occupants subjected to noise from two different activities. Internoise conference, Glasgow, 2022. 5. Zhang, Yuanyuan, Ou, Dayi, and Kang, Shengxian. The effects of masking sound and signal-to-

noise ratio on work performance in Chinese open-plan offices. Applied Acoustics , vol. 172, 2021. 6. Hui, Ct Justine, Ikuta, Moeto, Obata, Mochinobu, et al. Spatially target adaptive speech masking–

A pilot study on masking effect and annoyance. Applied Acoustics , vol. 182, 2021. 7. Zarei, Farideh, Lee, Joonhee, Mackenzie, Roderick, et al. Evaluation of the uniformity of sound-

masking systems in an open-plan office. Applied Acoustics , vol. 186, 2022.

8. Puyana-Romero, Virginia, Maffei, Luigi, Brambilla, Giovanni, et al. Sound Water Masking to

Match a Waterfront Soundscape with the Users’ Expectations: The Case Study of the Seafront in Naples, Italy. Sustainability , vol. 13, no 1, p. 371, 2021. 9. Philippe Sarnin and Marc-Eric Bobillier Chaumon. New forms of work and new business models.

Handbook of Work and Organizational Psychology: Psychological Issues at Work, chapter 5. De Boeck Supérieur , ISBN-13 97828 [In French], 2021. 10. Ignasi Capdevila. A typology of open innovation spaces based on the different modes of

innovation and motivations for participation [In French]. Gestion 2000, 33(4): p. 93–115, 2016.