Next Article in Journal
Synergistic Biochar–Nitrogen Application Enhances Soil Fertility and Compensates for Nutrient Deficiency, Improving Wheat Production in Calcareous Soil
Previous Article in Journal
Utilizing Farmers’ Views and Attitudes to Hinder Climate Change Threats: Insights from Greece
 
 
Article
Peer-Review Record

Assessing the Influence of Sustainability Using Artificial Neural Networks in Construction Projects

Sustainability 2025, 17(5), 2320; https://doi.org/10.3390/su17052320
by Manikandaprabhu Sundaramoorthy 1, Durgesh Kumar Sahu 1, Varadharajan R 2, Sudarsan Jayaraman Sethuraman 3,* and Ahmad Baghdadi 4,*
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3:
Reviewer 4: Anonymous
Sustainability 2025, 17(5), 2320; https://doi.org/10.3390/su17052320
Submission received: 9 December 2024 / Revised: 2 March 2025 / Accepted: 4 March 2025 / Published: 6 March 2025
(This article belongs to the Section Sustainable Engineering and Science)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article primarily explores the factors influencing sustainability in construction projects and utilizes the Artificial Neural Network (ANN) framework approach to enhance manageability, identifying and improving key factors affecting the sustainability of the construction industry. The article collects data through a questionnaire survey and organizes information using a qualitative meta-analysis method, preparing the questionnaire through literature research, case studies, and interviews. Subsequently, statistical analysis is used to explore the factors with the greatest impact on the sustainability of construction projects and compares the perspectives of different respondents. The experimental results reveal that job safety is the main factor affecting the sustainability of construction projects, while material usage and facilities, as well as internal and external challenges in the construction industry, are also significant factors affecting sustainability. The conclusion of the article emphasizes the effectiveness of ANN in predicting factors affecting the sustainability of the construction industry and points out that the Relative Importance Index (RII) analysis and ANN methods assess important factors impacting the sustainability of construction projects, providing in-depth insights for the sustainability of projects in the construction industry. However, there are still some shortcomings:

1. The article mentions relevant research on sustainable development and Artificial Neural Networks (ANN), but when discussing the factors of construction industry sustainability, it could further delve into the detailed research findings, differing viewpoints, and unresolved issues regarding these factors in existing literature.

2. The article's data collection is mainly completed through a questionnaire survey of construction and demolition sites in a specific region (Chennai, India), and the questionnaire is limited to certain personnel in the construction field. To enhance the universality and reliability of the research results, it is recommended to expand the geographical scope of data collection to cover construction projects in different regions; at the same time, increase the diversity of survey subjects, including related enterprises upstream and downstream in the construction industry, regulatory departments, and building users, to obtain more comprehensive and multi-angle data.

3. After deriving the research results, further discussions could be conducted on the practical significance and implications of these results for the sustainable development of the construction industry.

Author Response

Dear Reviewers

please find attached all of your valuable comments addressed and taken into consideration  

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript aims to assess factors influencing sustainability using Artificial Neural Networks (ANNs), but it fails to provide a clear and concise research question or hypothesis. The focus appears scattered across multiple unrelated aspects of sustainability. Below are the detailed comments:

1.       The author should justify the use of ANNs by comparing their performance to simpler methods and demonstrate their superiority for this context.

2.       The questionnaire with 84 questions appears overly lengthy and risks introducing response fatigue among participants, potentially compromising data quality. The Likert scale approach is poorly linked to the ANN analysis, with little explanation of its transformation into input variables. 

3.       With only 101 valid responses, the sample size is insufficient to generalize findings in a study involving ANN training and validation.

4.       The ANN model is evaluated based solely on R² and RMSE, with no external validation or sensitivity analysis. Please perform external validation, cross-validation, and sensitivity analysis to demonstrate the model's robustness and reliability.

5.       Discuss how the model can be generalized to other regions or industries, or acknowledge its limitations.

6.       The manuscript contains numerous grammatical errors, awkward phrasing, and inconsistent use of terminology, which hinder readability.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Author Response

Dear Reviewers

please find attached all of your valuable comments addressed and taken into consideration

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper explores the application of Artificial Neural Networks (ANNs) in assessing sustainability impacts within construction projects, focusing on sustainability issues in building projects, which constitute a significant topic of current importance in the construction industry and possess strong practical relevance. The research aims to identify key factors influencing the sustainability of construction projects and evaluate their relative importance, providing some references for the sustainable development of the construction industry. However, there are still several issues that require further clarification and refinement:

1. In the introduction, existing literature on sustainability in the construction industry is mentioned, but the distinctions and connections between this study and prior research are not clearly articulated. Additionally, when elucidating the innovative points, the paper fails to adequately demonstrate its uniqueness in terms of theory, methodology, or application. Although the paper attempts to apply ANNs to assess the factors influencing sustainability in the construction industry, this application is not entirely novel and lacks in-depth theoretical exploration and technological innovation.

2. To enhance the readability and persuasiveness of the paper, the author could include a detailed explanation of the working principle and output results of the ANN model. This could encompass the input and output variables of the model, the number of hidden layers and nodes, the hyperparameter settings during the training process, and so forth.

3. The characteristics of the collected samples have not been analyzed in detail, such as whether the response ratio among personnel of different positions is balanced, which may lead to biased results. It is recommended to adopt stratified sampling methods or similar approaches, sampling based on factors such as different types of construction projects, scales, and geographical locations. Additionally, conducting a cause analysis or appropriate follow-up on individuals who did not respond to the questionnaire should be performed to reduce sample bias.

4. The paper merely presents the results of the Relative Importance Index (RII) analysis and the ANN model in a simplified manner, without delving into the reasons for the differences between the two results and their intrinsic connections. For instance, job security ranks first in the RII analysis, but the unique influence mechanism of job security compared to other factors in the ANN model is not thoroughly analyzed. This prevents the readers from clearly understanding the comprehensive picture of how various factors impact sustainability.

5. For some factors (such as political issues and equipment maintenance), the Cronbach's alpha values are relatively low. The author should further explain whether these factors will have an impact on the overall results and how to handle these low-reliability factors.

6. There are some grammatical errors or inappropriate word choices in some expressions, for example, "are autonomously may not be so important" should be revised to "may not be so important independently." During the paper revision process, it is necessary to carefully check the grammar and wording to ensure accurate and clear expression.

7. It is recommended to discuss the limitations of this study and propose directions for future research.

Author Response

Dear Reviewers

please find attached all of your valuable comments addressed and taken into consideration

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

 

This manuscript introduces an enhancement in manageability to improve the factors that can actualize some imperative factors with the help of Artificial Neural Network (ANN) framework approach. The main topic is to discover the present variables that influence the supportability in the development industry. The proposed method further validates the accuracy of RII from another perspective. The ANN model was employed to predict and analyze the key factors influencing sustainability in construction projects, providing an additional layer of insight. The outcomes from the ANN model were compared with those derived from the RII analysis, and it was found that both methods identified similar key factors affecting sustainability, such as job security, material usage, and internal/external challenges in the construction industry. Although the ranking of these factors differed slightly between the two methods, the core findings remained consistent. This alignment between the ANN and RII results provides strong validation for the accuracy of the RII approach. The numerical studies of this work are solid, and clearly show the efficiency of the proposed method.

 

The proposed method is interesting, while the manuscript is very well written and easy to follow. I feel enjoyable to read it. Therefore, I suggest this manuscript to be accepted after a few revisions.

 

On line 5 of page 2, “different sort of variables” should be “different types of variables”.

 

On line 8 of page 2, “the most critical point which examined” should be “the most critical points examined”.

 

On line 9 of page 2, “material usage and facility, internal and external challenges in construction industry” should be “material usage and facilities, internal and external challenges in the construction industry”

On line 12 of page 2, “materials and others factors” should be “materials and other factors”.

 

On line 13 of page 2, “despite of administration life” should be “despite the administration life”.

 

On line 31 of page 2, “which is known as neurons” should be “which are known as neurons”.

 

On line 4 of page 3, “input layer, Hidden layer, and output layer” should be “input layer, hidden layer, and output layer”.

 

On line 12 of page 3, “Fly ash concrete promoting sustainable development by soft computing models which involves ANN and SVM used to predict fly ash concrete typr for controled concrete” should be “Fly ash concrete promotes sustainable development by soft computing models which involve ANN and SVM used to predict fly ash concrete type for controlled concrete”.

 

On line 18 of page 3, “at different construction and demolition site” should be “at different construction and demolition sites”.

 

On line 11 of page 5, “The Neural system tool kit of the MATLAB programming is used to make” should be “The Neural Network Toolbox of MATLAB”.

 

On line 40 of page 5, “Son et al. (2013) examine a model” should be “Son et al. (2013) examined a model”.

 

On line 12 of page 6, “It is also use for Predicting” should be “It is also used for Predicting”.

 

On line 14 of page 6, “The ANN used in high strength fibrous concrete which is subjected to elevated temperature” should be “The ANN is used in high- strength fibrous concrete which is subjected to elevated temperatures”

 

On line 10 of page 8, “the value of Mean Square Error, Root Mean Square Error” should be “the value of Mean Squared Error, Root Mean Squared Error”.

 

On line 15 of page 9, “Fig. 5 represent the performance of Neural Network tools” should be “Fig. 5 represents the performance of Neural Network tools”.

 

 

Author Response

Dear Reviewers

please find attached all of your valuable comments addressed and taken into consideration

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors just give rebuttal but without adequat response and revision. I do not think this paper can be published in Sustainability.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Author Response

Comments 1: The author should justify the use of ANNs by comparing their performance to simpler methods and demonstrate their superiority for this context.

Response 1:

Thank you for the valuable suggestion. Simpler methods may solve linear problems. While ANN will able to solve complex non-linear problems, with any number of data set having variation in the response. We have included the benefits and comparison of ANN over other methods. In future work, we can incorporate the other methods which can also solve the non linear problems and make the comparison.

Refer : Line 85, 110, 127, 214, 293

 

Comments 2: The questionnaire with 84 questions appears overly lengthy and risks introducing response fatigue among participants, potentially compromising data quality. The Likert scale approach is poorly linked to the ANN analysis, with little explanation of its transformation into input variables.

Response 2:

Thank you for the insightful suggestion. The question are prepared for the data required to perform the analysis. Limiting the question might lead to lack of data for the analysis. However considering the suggestion on response fatigue we will reduce the number of questions in the future studies. In addition, we can incorporate best feature selections techniques in the upcoming works. We can refine the questionaries for the future works.

Refer : Line 84

 

Comments 3: With only 101 valid responses, the sample size is insufficient to generalize findings in a study involving ANN training and validation

Response 3:

Thank you for the comments. The study was conducted in the Chennai region and limited to certain projects. Also it was conducted among selective category of employees or construction professionals. Hence there are limitations in the responder selection the responses are less.

 

Comments 4: The ANN model is evaluated based solely on R² and RMSE, with no external validation or sensitivity analysis. Please perform external validation, cross-validation, and sensitivity analysis to demonstrate the model's robustness and reliability.

Response 4 :

Thank you for the suggestions. Due to the limitations of data, we have applied the cross-validation methods to improve the evaluation score. In this research, we use k-fold validation techniques, whereas K=5.Due to this, each data appear at least one time in the testing process. Due to this, the average sensitivity score is listed in the Table 2.

 

Comments 5: Discuss how the model can be generalized to other regions or industries, or acknowledge its limitations.

Response 5:

Thank you for the response. The current method of analysis can be adopted to extended region may be for a country or globally. Accordingly, the data need to be collected and validated. However, the data might be diverse as there will be diverse in the factors such as culture, regulation, topography etc. Even though the limitation of the current study is acknowledged.

Refer : Line 397

 

Comments 6: The manuscript contains numerous grammatical errors, awkward phrasing, and inconsistent use of terminology, which hinder readability.

Response 6:

Thank you for the comment. The manuscript is completely revised with a proof reading for the language and the grammatical errors are removed. We use external sources to fine tune paper according to the English standards.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

Accept.

Author Response

Reviewer 1

The authors thank the reviewer for the comments and suggestions on the article. The corrections as suggested has been made and the response are given below

The article mentions relevant research on sustainable development and Artificial Neural Networks (ANN), but when discussing the factors of construction industry sustainability, it could further delve into the detailed research findings, differing viewpoints, and unresolved issues regarding these factors in existing literature.

We thank the reviewer for the valuable suggestion. More research finding related to the factors influencing sustainability is added to the article. Also other research methodology which focus on sustainability strategies are included.

Line 36, 48, 79

 

 

The article's data collection is mainly completed through a questionnaire survey of construction and demolition sites in a specific region (Chennai, India), and the questionnaire is limited to certain personnel in the construction field. To enhance the universality and reliability of the research results, it is recommended to expand the geographical scope of data collection to cover construction projects in different regions; at the same time, increase the diversity of survey subjects, including related enterprises upstream and downstream in the construction industry, regulatory departments, and building users, to obtain more comprehensive and multi-angle data.

Thank you for the suggestion. The region and the participants for the data collection is limited and acknowledged in the article.

This study is specifically focused on residential and commercial construction projects within the Chennai region. The scope was defined to ensure a detailed and context-specific analysis of sustainability factors affecting construction and demolition activities in this urban setting. By concentrating on these two project types, the study aims to assess sustainability challenges, policies, and industry practices that are most relevant to the local construction sector.

Furthermore, while the study primarily focuses on the Chennai region, future studies could conduct comparative analyses between different metropolitan areas, suburban developments, or rural construction projects to examine regional variations in sustainability practices. This would provide valuable insights into how sustainability strategies can be adapted to different economic and environmental contexts.

Line 62

 

 

After deriving the research results, further discussions could be conducted on the practical significance and implications of these results for the sustainable development of the construction industry.

Thank you for the suggestion. The result and discussion further expanded with the detailed elaboration of the result significance of the sustainable development in construction.

Line 343

Line 395

 

Back to TopTop