Using Regression Model to Develop Green Building Energy Simulation by BIM Tools

Reviewer1

Response

Introduction part: The literature and theoretical background should add more relevant studies in order to understand the current situation and serve as a precedent for the study to be carried out.

In the limited time available, we have reviewed as much literature as possible relating to green building. Please consider that this research was completed in about a year along with work on other projects. Therefore, we believe that all the basic information can be understood with the reading of this paper. The relevant theories and backgrounds are described in detail in the literature review.

Abbreviation section: The acronyms should be defined at first appearance in the manuscript and then must be consistently used throughout the manuscript. For instance, the term BIM is defined several times (L. 35, L.45, L. 150, L.151, L.153…)

The abbreviations are listed in a table in the Appendix section and the term BIM has only one explanation, Building Information Modelling, and is marked in the Abbreviations Indication.

Conclusions: It would be a good idea to merge subsections 5.2, 5.2 and sections 6 and 7. I think all this information can be considered in a single section called "Conclusions".

As a result of the researchers' discussions, we believe that a more detailed and precise discussion of an idea can be achieved by separating the 'conclusion' section. Separating the conclusion also facilitates the reader's guidance.

Bibliography:  Check the references and their style are according to the journal requirements. Please add the DOI of each investigation considered.

Every reference we make ensures compliance with the journal's requirements. DOI will not create a note in the reference because of the difference in reference style and the homogeneity of the format.

Specific comments:

I recommend changing the title of subsection 1.3 as it does not provide any information about the content and I would remove the title of subsection 2.1.

Comments reviewed

Reviewer2

Response

Introduction:
1. The literature is not reviewed well; I suggest the authors include the recently published articles (From 2019 to 2021) that also need to be cited in the revised version. The authors should clearly show what have we done and why this study is important in the introduction section.

Comments reviewed

2. I couldn't find the innovation of this research, what is the research gap, and why this study is important? Please elaborate.

There are research gaps in the following aspects: Limited research has been done on using BIM for energy simulation Energy modelling through Autodesk Revit and Green Building Studio needs to be further studied. A simplified method for developing a regression model needs to be proposed. The studies concocted on internal building shape are limited.

Methods:
1. Are twelve variables are enough and can provide a comprehensive analysis? Please elaborate.

Firstly, the twelve variables mentioned in this paper were selected for the energy analysis variables available in the Green Building Studio software. The variables were purposively selected for this study and regression analysis was carried out on those variables that could be analysed in the data. The reasons for this selection are explained in Table 2. These twelve variables are to some extent sufficient for a comprehensive energy analysis using the green building studio. However, many variables cannot be specifically quantified in real life. Therefore, the twelve variables are not comprehensive enough for a practical energy analysis.

2. I would suggest authors provide a detailed figure for the methodology.

Comments reviewed

3. What is the validation process which could explain that the data is correctly collected and comprehensively analyzed?

In this study, we firstly plot the scatter plot for those data. the data sets were firstly filtered through a series of analysis requirements. after we formulated an equation to predict energy, R studio regression analysis provided us the plot of residuals, t-test results, F-test results, standard error and correlation index. The data validation would based on those statistical index to approve that the process is correct. Overview of model criterion is also implemented.

Results:
1. Please provide enough details of codes and data in a supplementary file to allow the work to be reproduced by an independent researcher.

Comments reviewed

2. How the present study could help stakeholders to adopt sustainable mobility in cities?

This paper describes the use of the energy simulation tool Green Building Studio and the energy variables that can be changed by this tool. This provides future researchers with a direction for energy modelling. The paper also summarises the future directions and recommendations for energy modelling. The paper is intended to inspire stakeholders to think about and choose materials for sustainability in buildings.
 

3. The authors can also recommend some future research work.

The recommendations has been mentioned in section 6 recommendations.

Conclusion
1. What are the limitations of this study?

Limitations has been discussed in section 5.2.

Reviewer3

Response

There have been similar researches before, and the authors' contribution in this paper is not sufficiently pronounced.

Comments reviewed

The presented research results are not convincing, and the authors themselves point out the existence of certain problems in the results of their work.

Comments reviewed

Authors must clearly point out novelties in their research in relation to existing similar research.

In our literature research we have not found anyone using Green building studio (GBS) data to do parametric statistics analysis. This paper features a linear regression based on the novel cloud software GBS. Although similar topics have been studied, the software used, the design factors studied, and the methodology are not identical. Different assumptions can also lead to different conclusions.

The research methodology, methods of obtaining results and validation of the models need to be explained more clearly and concisely. The authors say that the regression models were validated using data for ABS buildings, but it is not explained what these buildings are nor is a comparison of the results shown. It is therefore not clear how the authors came to the conclusion of errors within 10%.

In this paper, the seven spatial models are first investigated and then the same energy simulation process is implemented for the spatial model of the ABS building. The energy simulation EUI of the ABS was compared with the EUI of the spatial model. The energy unit of EUIs is (MJ/m2/Year), m2 eliminates the effect of space, usage per year eliminates the effect of time and MJ determines the energy unit. Therefore the 7 spatial models and the ABS spatial model are comparable. The error within 10% is determined by the determination factor R square.

The analyzed ranges of values of the included building parameters must be clearly defined. The U-values of the wall, roof, window are nowhere exactly stated.

The exact range of U-values for walls, roofs and windows are indicated in Table 2. The analytical ranges for all building parameters are provided in the basic documentation of the green building studio. reference includes the source of the basic settings for the energy simulation.

The impact of individual parameters on energy consumption should be investigated in more detail and physically explained, and appropriate conclusions should be defined on that basis.

Comments reviewed

At the end of the paper, the authors give general recommendations for the implementation of buildings energy consumption optimization, which either do not follow from the research results or they themselves did not follow that recommendations in their work.

Since only one energy software was used in this study, limitation was not possible to eliminate completely. The final recommendations of the paper include both reflections on this study and suggestions for other future researchers.

Reviewers’ comments

Authors’ Reply

Figures should be provided with high resolution?

We have done our best to provide high resolutions Figures. These figures were originally developed in R-studio.

Please provide enough details of codes and data in a supplementary file to allow the work to be reproduced by an independent researcher

The relevant supplementary materials were provided.

Reviewer3

Response

1

In my opinion, the authors made some changes, but did not answer the questions asked in an appropriate and expected way. That is why I still believe that the work in its current form is not suitable for publication.

Comments reviewed

2

The authors still do not give a clear answer to the question of what is new in their paper compared to the similar studies, which they must do not only in their comments to reviewers but also in the content of their manuscript.

     This is research that is totally related to building technology.  It is not just covered architectural approaches. The main contribution of the current research is to undertake a quantitative and qualitative analysis as well as comprehensively evaluate some critical parameters which are classified as significant factors in the design of modern buildings.

The main novelty of this study is to select a particular case study (ABS building in Sydney – Australia).  All the important critical design factors, which are considered to design building technology approaches in the former studies, are systematically examined in the ABAS building.

Besides, the main aims of this research are in line with the recently published paper titled “Implementation of BIM Energy Analysis and Monte Carlo Simulation for Estimating Building Energy Performance Based on Regression Approach: A Case Study”.

1. We performed a multilinear regression on the WWR. The dropping method was used in this regression analysis. Variables that had little or no effect on the WWR were excluded from the current regression. 2.  We further investigated the relationship between coefficient and constant for design variables in Linear Regression Models. We further investigated the relationship between coefficient and constant for design variables in Linear Regression Models. The design variables for which shape has a large impact on energy are summarised.

3

The literature review included in the section 6 Recommendation should be refined and those studies that concern the authors primary goal summarized in the sentence "The paper aims to identify design parameters critical to BEP to assist architects in the initial stages of building design and investigate their relationship" moved to the section 2 Literature Review. At the end of section 2 the authors should emphasize their contribution in relation to already published papers. Using different software alone is not enough of a novelty.

"The paper aims to…their relationship "was moved to abstract. Section 2.7 was added to summarise author's contribution.

4

It is still not clear what ABS buildings are (at least perhaps to non-architects). The authors state "The error within 10% is determined by the determination factor R square." Where is that shown in the manuscript?

Section 3.3 was added in to introduce ABS building. The 10% error was delated. We decided not to discuss this point here.

5

The authors comment "The exact range of U-values for walls, roofs and windows are indicated in Table 2. The analytical ranges for all building parameters are provided in the basic documentation of the green building studio. The reference includes the source of the basic settings for the energy simulation. ". I have expected the authors to include in their manuscript the range of analyzed values of heat transfer coefficients, ie. (from-to) in W/(m2K). This is not contained in Table 2.

Table 6 contains material Thermal Properties, as required by reviewer3.

6

 Since the aim of the author was to determine the design parameters critical for BEP, it would be desirable in the last sections of the manuscript to compare the findings with other studies. Crucial design parameters should basically be the same regardless of the software used.

Comments reviewed

The third Reviewer’s comments

Authors’ Reply

The authors have so far not made it known that they presented a part of the study in another paper. The published paper provides answers to some previous questions. The authors must cite their previous paper in the current manuscript.

Looking now at the study as a whole, new questions arise. If the authors wanted to compare the energy performances of different building shapes, they had to keep the same basic characteristics of the building (volume of the conditioned space, size and number of the windows, orientation, etc.). I am not sure they did that, but it is clear now that the spaces they used to validate their regression models do not have the same characteristics. The volumes of the spaces, the window-to-wall ratios, and the orientation seem to vary, so I do not think it makes sense to compare energy consumption the way the authors did to determine which shape of the building is most energy efficient.

If we judge based on the regression models (Figures 12 and 13), and if the basic characteristics of the analyzed building shapes (Figure 3) are identical (must be specified in the manuscript; to keep just the same floor area is not enough), I do not think it can be concluded that triangle-shaped buildings are most energy efficient and diamond-shaped buildings are most energy inefficient.

The previously published paper was cited.

Unfortunately, you have changed your mind and made new questions and comments. By respecting your approach, in order to address the new comments, a new study should be undertaken. I would like to deeply appreciate your new scopes of the current research; however, it needs to prepare a new paper to reply to your new questions. Thus, we are unable to fully cover all the new comments.

Also, regarding comparing triangle-shaped buildings and diamond-shaped buildings, our conclusion is just a hypothesis and further experimental investigations are required.

Minor comments:

The authors were asked to write the considered ranges of U-values in W/(m2K). The U-value ranges for the walls and the roof are (should be) presented in Figures 16 and 17 in the previously published paper. Table 6 is not required if the U-values ​​of the building elements change during the simulations.

The abstract still states “the developed regression models are validated to within 10% error via a case study of the ABS building.” Later in the paper, this was omitted.

Line 43: “should” instead of “Fshould”

Lines 142, 148, 154: “MLR” instead of “MRL”

Lines 180, 181, 182: Please insert a sentence between the section title and Table 1.

Section 2.7: I suppose more than one person contributed to the research (researcher/s)

Table 2, infiltration rate: the stated values should be expressed in l/s/m3. However, the values do not represent air changes per hour but specific air flow rates.

Please check that every figure in the manuscript is linked with (mentioned in) the text. Some of the figures are just inserted.

The list of abbreviations could be included in a different way, not as an appendix.

All of the Minor comments were addressed.