Comfort Distance—A Single-Number Quantity Describing Spatial Attenuation in Open-Plan Offices

Round 1

Reviewer 1 Report

The manuscript aims to provide the basis for comfort distance as a viable single number quantity (SNQ) in room acoustic measurements in offices using ISO 3382-3. It does so by briefly outlining the context of this SNQ, its relationship with existing SNQ, and follows that up by investigating and justifying its criterion level. They have also proposed a ranked classification scheme for the ISO 3382-3 SNQs and comfort distance. The method uses data primarily from one of authors’ previous studies since it has the largest range for ISO 3382-3 SNQs. Other studies that have measured several offices using ISO 3382-3 have also been summarized and used to confirm the findings.

The paper is concise, devoid of major issues, and will most likely assist the ongoing update of ISO 3382-3. The methods used are fairly straightforward and seem sufficient. More advanced methods could have been used, but it seems highly unlikely that they would have returned substantially different results. The discussion and conclusions are terse and assist the overall readability.

While the following comments are mostly minor, the main criticism is the presence of some speculations in parts of the introduction (highlighted below), which is unnecessary and should be fixed to lean towards an objective appraisal, to reflect the scientific tone of the rest of paper.     

Specific comments:

L13-14: Please change this sentence to something like ‘The definitions of both quantities may sometimes be difficult for a non-technical audience’. The emphasis is on the ‘may be’. In the experience of this reviewer, non-acousticians have actually been quite receptive of these SNQs and have no trouble in understanding them. This is arbitrary too, but at least counters the authors’ experience. Besides, comfort distance can exist regardless of whether people can understand the other SNQs.

L28: Would be better to say, 'One of the ways to control the…’. or similar. Otherwise, without reading further, it sounds like it is the only way.

L32: Change to something like ‘ ….office alone cannot solve …’

L34: Consider ‘beyond’ instead of ‘above’

L34-35: ‘Therefore … noise’. Please change to something like, ‘In such cases, and also in general, room acoustic means can be used for noise control’.

L36: Instead of ‘room attenuation’, might be better to say 'attenuation of room reflections'

L37: Consider ‘electroacoustic sound masking’

L44: ‘…room acoustic performance…’. To remember that we are just looking at unoccupied spaces

L67: ‘measure’ to ‘measuring’

L91: Instead of ‘big’, use ‘bigger’ or ‘general’.

L100-107: Most of this paragraph reads very anecdotal and without much evidence, which doesn’t look good in an otherwise solid scientific manuscript, and also in general. Either remove the judgment-heavy sentences, or heavily pad them at relevant points with 'in the authors' experience', 'in our opinion', etc.

As stated above, in the experience of this reviewer, more non-acousticians have understood and actually embraced these metrics than not.

The previous paragraph also has a similar tone, though it at least includes references to almost provide some balance.

L101-102: Preferable remove ‘The main … [dB]’, or justify it better without speculating.

L119,122: Please put equation numbers.

Fig.2: The legend looks useless in its present form. Either change grayscale intensity, etc., or consider removing the legend.  

L153: 'Fig 3'

L179: ‘analysis of how…’

L182-184: It might be worth clearly specifying the lower and upper limits to avoid misuse. For instance, rC of 7 m could either go to class B or C currently.

L197-198: ‘The SPL … lower’. Cite the sources. In many cases, low frequency fluctuations in the ventilation noise can be uncomfortable even at low levels, which might be worth pointing out.

L225-226: ‘Final … decision’. This may be true (again, anecdotal), but is there really a need to say this?

 

Author Response

See attachment

Author Response File: Author Response.docx

Reviewer 2 Report

Dear Authors, I really appreciated the study that you present to Applied Sciences that describes a temptative to substitute two more "complicated" quantities included in ISO 3382-3 with a simpler one, that is the comfort distance.

I understand the rationale that is under the choice of the distance and its reference valued of 45 dB(A) for the speech level, but it is not clear to the reader because the comfort distance, which does not include noise level in its calculation, should be preferred to the distraction distance or to the privacy distance, that are based on STI that have been understood already by non-acousticians, as you stated in the introduction. STI does include both speech and noise level at the receiver postion. Measurements should be done with different possible noise levels that depend on the country. Please comment on that.

Line 206 pg. 8: "The mean levels during the workday are usually 48−59 206 dB LAeq,8h according to Yadav et al. [23]. This supports also that 48 dB might not be comfortable since it exceeds the average activity noise level." Why do not consider 48 dB for rC? 45 dB is not reliable in real offices. Moreover it depends on the latitute of the office. In southern Europe noise is much more high than in officies in northern europe that constitute the base of your investigation.

Line 2010, pg. 8: "Thus, setting the comfort criterion level higher than 45 dB is not supported from this practical viewpoint." What does it mean? It works with 48 dB.

Please number the equations.

Line 170, pg. 6: "The comfort distances of the 26 offices of Ref. [14] for comfort criterion levels ranging from 30 to 50 dB are shown in Fig. 4." Comfort statistics.

Line 171, pg. 6: "The corresponding probabilities that the comfort
distances exceeded the length of the office are shown in Fig. 5." How did you obtain the probability. Please explain.

Line 187, pg. 6: "The range of comfort distance values, rC, as a function of the comfort criterion level, Lp,A,C, for the 26 offices of Ref. [14] calculated by Eq. (1). Mean, maximum, and minimum values within the sample of 26 offices are shown." Eq. (2).

Author Response

See attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This manuscript presents the scientific background supporting the introduction of comfort distance (rC) as a new SNQ in the revised version of ISO 3382-3 (at the stage of DIS as of May 2021). From data obtained from previous studies, typical comfort distance values are calculated and limit values for a possible classification are proposed.

The paper is well-organized, interesting and easy to read. However, I would recommend some more explanations or considerations in specific points of the manuscript to improve the clarity and completeness of the paper to the general reader; moreover some typos should be fixed. Here under some suggestions to the authors.

• The Introduction gives interesting insights on the development of ISO 3382-3:2012, however it is not mentioned the fact that this standard allows using two methods (called direct and indirect methods) to measure D2,S. It is important to recall that in presence of a high level of HVAC noise during measurements the direct method is more robust and should be preferred. See for example:

D’Orazio et al. (2018). Comparison of different in situ measurements techniques of intelligibility in an open plan office, Building Acoustics 25(2) 111-122.

This consideration should complement the one in lines 55-56 of the manuscript.

• Figure 1. In the legend, it should be clarified from which studies the data (black dots) come from.
• Equation (1) (line 11) and equation (2) (line 122) are not numbered. Please do.
• At the end of the Introduction, (lines 131-132) it should be recalled that on one hand the comfort distance has been introduced in the new version of ISO 3383-2:2021 and on the other hand the privacy distance rP has been dropped.
• Table 1. Why some offices are identified with numbers (1, 2, …) and others with letters (A, B, …)? Please use a uniform labeling.
• Legend of figure 2, line 148: what “The number of measurement distances per office ranged from 4 to 13” does exactly mean? Please clarify.
• Line 153: “Fig. 2 for office ID 1” should read “Fig. 3 for office ID 1”.
• Lines 173-174. Instead of saying “The following results already assume that the comfort criterion level is Lp,A,C=45 dB as our later discussion gives justification for this choice”, it could be better to say that Figure 5 clearly shows that the probability P drops to zero when Lp,A,C > 45 dB and thus this value is taken as a criterion, as the following discussion confirms.
• Line 218: “the mean value of rC was 8.7 dB” should read ““the mean value of rC was 8.7 m”.
• Line 226: “Final target values or classification limits of are political decisions” should read “Final target values or classification limits are political decisions”.
• Conclusions, lines 231-235. On one hand it is said that “The mean value of comfort distance was 9 meters in our database”, one the other hand it is said “The worst class (D) reached is when comfort distance is between 9−11 m”. It sems unbalanced to set the worst class limit close to the mean value (9 m). This point needs a clarification or a revision of the limit values of the classes.

In my opinion this work can be accepted after a minor revision to fix the above issues.

Author Response

See attachment.

Author Response File: Author Response.docx