Machine Learning Models of Exergoenvironmental Damages and Emissions Social Cost for Mushroom Production

Round 1

Reviewer 1 Report (New Reviewer)

The article contains a valuable report of research conducted with the use of artificial intelligence. It is of great cognitive, practical and scientific importance. In my opinion this paper should be published in Agronomy. Please take into account (when improving the article) my following comments:

1. The abstract doesn’t require changes.

2. The tables in the introduction contains valuable and information carrier. However, I don’t see a desciptions related to studies carried out by a few specific researchers (from this table). Complete it, please.

3. Improve the visibility of figures 3, 5 and 6.

4. Larger font unify the font in whole of the text.

5. Point 2.2.1 – the content doesn’t have to be included in the form of a separate chapter or change the title of this chapter, please - it is misleading.

6. Complete literature reports in your discussion.

Author Response

Agronomy

Machine learning models of exergoenvironmental damages and emissions social cost for mushroom production

Dear Editor-in-Chief

Thank you for your valuable comments and suggestions on the structure of our manuscript. We have modified the manuscript accordingly and the detailed corrections are added to the manuscript in red. Also, the detailed corrections are listed as follows:

Reviewers' comments:

Reviewer 1

The article contains a valuable report of research conducted with the use of artificial intelligence. It is of great cognitive, practical and scientific importance. In my opinion this paper should be published in Agronomy. Please take into account (when improving the article) my following comments:

Thanks for your positive feedback. We investigated your valuable comments and applied all of them to avoid any ambiguity.

1. The abstract doesn’t require changes.

Thanks for your comment.

2. The tables in the introduction contains valuable and information carrier. However, I don’t see a desciptions related to studies carried out by a few specific researchers (from this table). Complete it, please.

Thanks for your comment. We rechecked this table again and we compeleted description. Of course, in several items, they had not do specific action in the mentioned item (column title).

3. Improve the visibility of figures 3, 5 and 6.

Thanks for your comment. We prepared the high-quality versions of all figures. After acceptance, we will consult with the technical progress to place insert them. In this moment, with respect to template size, we had to insert the figures with this size.

4. Larger font unify the font in whole of the text.

Thanks for your comment. We modified whole of manuscript.

5. Point 2.2.1 – the content doesn’t have to be included in the form of a separate chapter or change the title of this chapter, please - it is misleading.

Thanks for your comment. We removed the number and rearranged the number of subsections.

6. Complete literature reports in your discussion.

Thanks for your comment. We modified discussion section based on the all-reviewer comments.

Best Regards

Author Response File: Author Response.docx

Reviewer 2 Report (New Reviewer)

Keyword: line 31. Please remove any words form this section which are used in the title.

1.       Introduction. Line 34-129. Please reword for common English usage, sentence structure, and grammar.

Line 34-129. The Introduction section is not clear and is very choppy. It does not provide a clear direction. It needs to be rewritten with a clear aspect as to what the paper/manuscript is actually about. Please rewrite and shorten.

Page 2-3 Line 54-55. Nomenclature section not really helpful as currently displayed in manuscript and unclear if it is actually necessary? Could the abbreviations be included in () after the first usage of each word in each section of the manuscript instead?

2.       Materials and Methods. Line 130-349. This section is overly long and needs to be shortened considerably. Please also reword for common English usage, sentence structure, grammar, etc.

3.       Results. Line 350-518. This section is also overly long and needs to be shortened considerably. Please also reword for common English usage, sentence structure, grammar, etc.

4.       Discussion. Line 519-603.This section should be combined with the results section for clarity. This section can be shortened as well.

The manuscript is interesting, but is quite hard to follow and quite choppy at times. It does not flow together. I feel it would benefit from some further work and refinement. 

Author Response

Agronomy

Machine learning models of exergoenvironmental damages and emissions social cost for mushroom production

Dear Editor-in-Chief

Thank you for your valuable comments and suggestions on the structure of our manuscript. We have modified the manuscript accordingly and the detailed corrections are added to the manuscript in red. Also, the detailed corrections are listed as follows:

Reviewers' comments:

Reviewer 2

Keyword: line 31. Please remove any words form this section which are used in the title.

Thanks for your comment. We revised this section according to your comment. The changes in the text were identified in red.

1. Introduction. Line 34-129. Please reword for common English usage, sentence structure, and grammar.

The text of the paper was examined in terms of the language structure of the English language and the changes in the text were identified in red.

Line 34-129. The Introduction section is not clear and is very choppy. It does not provide a clear direction. It needs to be rewritten with a clear aspect as to what the paper/manuscript is actually about. Please rewrite and shorten.

Thanks for your comment. We rewroted and shortened the introduction section according to your comment. Also, a paragraph was added to the introduction to avoid any ambiguity. The changes in the text were identified in red.

Page 2-3 Line 54-55. Nomenclature section not really helpful as currently displayed in manuscript and unclear if it is actually necessary? Could the abbreviations be included in () after the first usage of each word in each section of the manuscript instead?

Thanks for your comment. We removed Nomenclature section according to reviewers comments. Also the abbreviations was added to manuscript inside () after the first usage of each word.

2. Materials and Methods. Line 130-349. This section is overly long and needs to be shortened considerably. Please also reword for common English usage, sentence structure, grammar, etc.

Thanks for your comment. The Materials and Methods section rewrote and shortened according to reviewers' comments. Some Figures and Tables were removed according to reviewers' comment. Also, the text of this section was checked in terms of the language structure of the English language The changes in the text were identified in red.

3. Results. Line 350-518. This section is also overly long and needs to be shortened considerably. Please also reword for common English usage, sentence structure, grammar, etc.

Thanks for your comment. The results section rewrote and shortened according to reviewers' comments. Table 5 was also modified to be shorter and more standardized. Also, the text of this section was checked in terms of the language structure of the English language The changes in the text were identified in red.

4. Discussion. Line 519-603.This section should be combined with the results section for clarity. This section can be shortened as well.

Thanks for your comment. The discussion section shortened and combined with results section according to your comment.

The manuscript is interesting, but is quite hard to follow and quite choppy at times. It does not flow together. I feel it would benefit from some further work and refinement.

Thanks for your favor. We revised whole manuscript according to reviewers’ comments and added some sentences to improve manuscript and avoid any ambiguity. The changes in the text were identified in red.

Best Regards

Author Response File: Author Response.docx

Reviewer 3 Report (New Reviewer)

Your study strikes me as conceptually very confusing, generic, and lacking a clear innovative contribution. 

On page 2, all that nomenclature is unnecessary in my opinion. In fact, acronyms should be written as they occur in the text and not separately (at least not in MDPI journals).

The Life Cycle Assessment is not spelled that way. In fact, you list all the inputs and outputs involved in the process, whereas you only list the emissions. Tables 2 and 3 should be eliminated, merged and summarised, without specifying the individual emission. You should see some literature articles on LCA to see how to write an LCI.

LCIA: What methodology did you use? Midpoint? Endpoints? Figure 3 has no scientific value, nor does Figure 4, although it is very beautiful and colourful. Those things can also simply be listed. 

Table 5 should be combined with Table 2 and 3, although it does not go between results and discussions.

Line 357: Use scientific notation.

Figure 7 and 8 are practically the same thing (characterised and normalised results). Choose one.

The conclusions are generic and scholastic, without a clear scientific contribution. In any case, the article is too long.

Author Response

Agronomy

Machine learning models of exergoenvironmental damages and emissions social cost for mushroom production

Dear Editor-in-Chief

Thank you for your valuable comments and suggestions on the structure of our manuscript. We have modified the manuscript accordingly and the detailed corrections are added to the manuscript in red. Also, the detailed corrections are listed as follows:

Reviewers' comments:

Reviewer 3

Your study strikes me as conceptually very confusing, generic, and lacking a clear innovative contribution. 

Thanks for your comment. We modified the structure of the article based on the reviewers' comments. The introduction section rewrote and shortened according to reviewers' comment. Also, a paragraph was added to the introduction to avoid any ambiguity. Some Figures and Tables were removed according to reviewers' comment. Discussion section combined with the results section for clarity.

On page 2, all that nomenclature is unnecessary in my opinion. In fact, acronyms should be written as they occur in the text and not separately (at least not in MDPI journals).

Thanks for your comment. We removed Nomenclature section according to reviewers’ comments. Also, the abbreviations were added to manuscript inside () after the first usage of each word.

The Life Cycle Assessment is not spelled that way. In fact, you list all the inputs and outputs involved in the process, whereas you only list the emissions. Tables 2 and 3 should be eliminated, merged and summarized, without specifying the individual emission. You should see some literature articles on LCA to see how to write an LCI.

Thanks for your comment. We summarized this section and removed Table 2 and 3 according to your comment.

LCIA: What methodology did you use? Midpoint? Endpoints? Figure 3 has no scientific value, nor does Figure 4, although it is very beautiful and colorful. Those things can also simply be listed. 

Thanks for your comment. We removed these Figures according to your comment.

Table 5 should be combined with Table 2 and 3, although it does not go between results and discussions.

Thanks for your comment. We removed Table 2 and 3 according to your comment. Table 5 was also modified to be shorter and more standardized.

Line 357: Use scientific notation.

This was corrected in the manuscript.

Figure 7 and 8 are practically the same thing (characterized and normalized results). Choose one.

Thanks for your comment. We removed Figure 8 according to your comment.

The conclusions are generic and scholastic, without a clear scientific contribution. In any case, the article is too long.

Thanks for your comment. We rewrote the whole article and removed some parts, Tables and figures to make the article shorter. Also, discussion section rewrote and combined with the results section for clarity.

Best Regards

Author Response File: Author Response.docx

Reviewer 4 Report (New Reviewer)

This paper evaluates intelligent models for exergoenvironmental damage and emissions social cost (ESC) for mushroom production in Iran by three machine learning (ML) methods, including adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN) and support vector regression (SVR). Environmental life cycle damages, cumulative exergy demand and ESC are examined by ReCiPe2016 method for 100 ton of mushroom production after data collection by interview. This paper has a great innovation in research methods.

The writing of this paper is standardized and rigorous, and no major problems have been found. However, there are still some questions I would like to discuss with the author.

   1.In the section of introduction, it is suggested that the author not only discuss the method, but also strengthen the explanation of the research problem and its significance.

   2. in the section of Materials and Methods, the author wrote that “Hence, data collected for the past 133 studies are used to perform the present research. Isfahan”, I wonder how the data was collected in the past. Moreover, since the “the past studies” were published in 2014-2015, could the data still be used for the present research?  Will environmental damage be mitigated, especially given the current improvements in equipment and technology?

3.In figure3: Whether land should be classified as resources?

Author Response

Agronomy

Machine learning models of exergoenvironmental damages and emissions social cost for mushroom production

Dear Editor-in-Chief

Thank you for your valuable comments and suggestions on the structure of our manuscript. We have modified the manuscript accordingly and the detailed corrections are added to the manuscript in red. Also, the detailed corrections are listed as follows:

Reviewers' comments:

Reviewer 4

This paper evaluates intelligent models for exergoenvironmental damage and emissions social cost (ESC) for mushroom production in Iran by three machine learning (ML) methods, including adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN) and support vector regression (SVR). Environmental life cycle damages, cumulative exergy demand and ESC are examined by ReCiPe2016 method for 100 ton of mushroom production after data collection by interview. This paper has a great innovation in research methods.

The writing of this paper is standardized and rigorous, and no major problems have been found. However, there are still some questions I would like to discuss with the author.

Thanks for your favor.

1. In the section of introduction, it is suggested that the author not only discuss the method, but also strengthen the explanation of the research problem and its significance.

Thanks for your comment. We rewrote and shortened the introduction section according to your comment. Also, a paragraph was added to the introduction to avoid any ambiguity. The changes in the text were identified in red.

2. In the section of Materials and Methods, the author wrote that “Hence, data collected for the past studies are used to perform the present research. Isfahan”, I wonder how the data was collected in the past. Moreover, since the “the past studies” were published in 2014-2015, could the data still be used for the present research?  Will environmental damage be mitigated, especially given the current improvements in equipment and technology?

Thanks for your comment. Given that the technology of mushroom production has not changed during this period. So, we used the data of the previous study.

3. In figure3: Whether land should be classified as resources?

Thanks for your comment. This figure was removed according to the recommendations of the reviewers. However, this figure is based on reliable references.

Best Regards

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report (New Reviewer)

N/A

Reviewer 3 Report (New Reviewer)

The authors adequately met the reviewer's requirements. The paper may be published.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The purpose of this article is to assess the effectiveness of three different machine learning methods - adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN), and support vector regression (SVR) - in predicting exergoenvironmental damage and emissions social cost (ESC) for mushroom production in Isfahan province, Iran. The results obtained from this study are valuable and add contributions to the existing literature. Further, this study has several tables and figures which are good enough and explained the results very well. The research and formatting are within the scope of the journal. However, I have a few comments：

1. The title of the article does not have a strong relationship with the content, and the content of the article does not point towards a common goal (the title). Specifically, the conclusion about mushroom cultivation is almost a separate part from the discussion on the strengths and weaknesses of machine learning models. From the perspective of the title, it seems that the focus of the article should be on machine learning models or research methods, but the majority of the content and important conclusions are not related to machine learning models. Please adjust the title or content to make the article more closely related to the title.

2. Machine learning models should make contributions to the analysis of real-world problems, rather than just comparing the strengths and weaknesses of methods in terms of statistics.

3. Please add annotations to the first figure to make the figure more clear and intuitive in reflecting your research content.

4. In Section 3.7, it would be best to do a comprehensive comparison of these methods from multiple dimensions using a series of existing research.

5. Please adjust equation (4) to make it more standardized and readable.

Author Response

Reviewer 1:

The purpose of this article is to assess the effectiveness of three different machine learning methods - adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN), and support vector regression (SVR) - in predicting exergoenvironmental damage and emissions social cost (ESC) for mushroom production in Isfahan province, Iran. The results obtained from this study are valuable and add contributions to the existing literature. Further, this study has several tables and figures which are good enough and explained the results very well. The research and formatting are within the scope of the journal. However, I have a few comments:

Thank you for your favor.

1. The title of the article does not have a strong relationship with the content, and the content of the article does not point towards a common goal (the title). Specifically, the conclusion about mushroom cultivation is almost a separate part from the discussion on the strengths and weaknesses of machine learning models. From the perspective of the title, it seems that the focus of the article should be on machine learning models or research methods, but the majority of the content and important conclusions are not related to machine learning models. Please adjust the title or content to make the article more closely related to the title.

Thanks for your comment. As mentioned in the content of the manuscript, the main goal of the article is to provide accurate models for predicting the environmental effects of mushroom production. But before this, all dependent and independent variables must be identified and calculated. Therefore, firstly, the method of calculating these parameters is discussed, and then predictive models are presented. However, we modified the title according to your recommendation. If you have any suggestions in this regard, we will be happy to provide them to us.

Life cycle assessment and machine learning models of exergoenvironmental damages and emissions social cost for mushroom production

2. Machine learning models should make contributions to the analysis of real-world problems, rather than just comparing the strengths and weaknesses of methods in terms of statistics.

Thanks for your comment.

In agricultural productions, it is necessary to study energy consumption patterns to select a sustainable and optimal model, reduce production costs and environmental pollution. So one of the momentous steps to meet sustainable agriculture goals is determining the relationship between outputs and inputs in agricultural production processes. Models can predict the environmental effects of agricultural systems, therefore, they are considered as an important tool for best management. In this regard, the more accurate the model is, the better the results will be.

Therefore, the first step in managing energy consumption and reducing the adverse environmental effects resulting from its consumption is to provide an accurate model to identify the influencing factors in the production process. In fact, this issue was the main purpose of this study.

3. Please add annotations to the first figure to make the figure more clear and intuitive in reflecting your research content.

Thanks for your comment. We modified the mentioned figure according to your advice and the changes are added to the manuscript.

4. In Section 3.7, it would be best to do a comprehensive comparison of these methods from multiple dimensions using a series of existing research.

Thanks for your comment. We modified this section according to your advice and the following red text are added to the manuscript to complete this section.

In the last section, the accuracies of different ML models are evaluated using statistical indicators. The results indicated that SVR model has the best performance in terms of all statistical indicators. Fig. 12 shows R² in different ML models, which indicates that R² of SVR model for predicting dependent variables including TWD, TCD and TESC are higher than R² of other models. In other words, SVR model outperforms others models. The reason for this is that in agricultural and environmental issues we are faced with non-linear systems, and since SVR model has outstanding learning, modeling with that has better results. Taheri et al. [66] applied ANN and SVR to model the drying of lentil in a microwave fluidized bed. The results indicated that moisture ratio and temperature of lentil could be predicted accurately using ANN and SVR. Performance evaluation of the models with statistical parameters indicated that ANN provided relatively better accuracy for prediction of lentil temperature. In another study Ghasemi-Mobtaker et al. [99] used ML models to predict output energy, economic profit, and global warming potential of wheat production. The results showed that ANN and ANFIS models outperform linear models to predict mentioned parameters.

5. Please adjust equation (4) to make it more standardized and readable.

Thanks for your comment. We adjusted the Eq. (4) to avoid any ambiguity.

Author Response File: Author Response.docx

Reviewer 2 Report

Dear Authors,

The main problem with the article is the lack of any context and relevance of the topic. The calculations made seem to be made for the sake of calculation itself. Also, from the point of view of using methods, the article could be excellent. 5 scientific methods are used, each of which could be worthy of an article. At the same time, all together they create an unclear structure of the article with weak reasoning, poor discussion, poorly connected to the results.

The article needs a strong topicality and justification of the topic, one goal and the tasks to achieve it, a discussion about the importance of using methods, the characteristics of the data used, improving the quality of the images.

In addition, explaining the abbreviations in the text, explain figure 2, I assume that in English it is The Ministry of Energy not Power Ministry of Iran, describe the weighting process for LCA. The novelty mentioned in the conclusions are easier to justify if there is a discussion about the previously conducted research.

Author Response

Reviewer 2:

The main problem with the article is the lack of any context and relevance of the topic. The calculations made seem to be made for the sake of calculation itself. Also, from the point of view of using methods, the article could be excellent. 5 scientific methods are used, each of which could be worthy of an article. At the same time, all together they create an unclear structure of the article with weak reasoning, poor discussion, poorly connected to the results.

The article needs a strong topicality and justification of the topic, one goal and the tasks to achieve it, a discussion about the importance of using methods, the characteristics of the data used, improving the quality of the images.

In addition, explaining the abbreviations in the text, explain figure 2, I assume that in English it is The Ministry of Energy not Power Ministry of Iran, describe the weighting process for LCA. The novelty mentioned in the conclusions are easier to justify if there is a discussion about the previously conducted research.

Thanks for your comment. I would like to inform you that the following changes were made according to your comments in the text of the article:

• All abbreviations mentioned in the text are listed in Nomenclature.
• Figure 2 was modified based on the comments of the reviewers.
• The Ministry of Energy was used in manuscript.
• The weighting process for LCA was described and the following red text are added to the manuscript to complete this section:

Weighting as a final and optional phase in life LCIA is subjective and implies a value judgment, which may influence the results of LCA. In this section, each environmental effect is weighed based on its performance in harming the environment. This step aims to determine the significance of each category and how important it is relative to the others. Each impact group that has more damage efficiency has a higher value. During weighing, various environmental impacts are weighed compared with each other. Weighting includes multiplying a weighting factor with the normalized results of each of the impact categories and thus shows the relative importance of the impact category [68]. According to its practicability for comparing impacts of different scenarios or products, supporting decision-making and result communication, it is commonly applied in research.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Dear Authors,

You use diverse and complex methods to establish obvious things about agricultural production. They are superficially described in the discussion section, and you seem to be the first to point this out, as there is no reference to other studies. The subject of your article is mathematical methods, not agricultural production. On the one hand, if the object of the article is agricultural production, then the article is unfounded and the conclusions general. On the other hand, if the object of the article is mathematical methods, then the article is superficially presented and lacks data to confirm the approach. At the same time, it must be recognized that the work has used a large number of citations (many self-citations) and invested a lot of work. It is not an integrated study, but different examples of the application of methods.

Author Response

Dear Reviewer

Thank you for your valuable time spent reading the article. All the answers to all your comments are provided as the following:

1- Trials and errors in applying inputs changes of agriculture crops require a lot of costs and time, and it is practically not possible for the farmers of developing countries like Iran, who mostly have monoculture. Therefore, machine learning methods can provide applicable sustainability models for further development without losing costs and times.

2- In connection with the superficial explanations as you told, whether in the agricultural sector or in the mathematical sector, first of all, the explanations are very deep, all the methods are presented in a complete and understandable way for the audience of the journal (especially agricultural experts). Secondly, irregular explaining the mathematical aspect not only outs the article from the scope of Agronomy and does not create an attraction for the audience, but also makes the volume of the article inappropriately out of standard. This article aims to compare machine learning methods in agricultural-environmental modeling with comprehensive aspects and has done this well. So that from the lowest level to the highest level, the audience can benefit from its results.

3- Valuable studies have been done in the field of modeling in agricultural production. A long list of them is presented in Table 1, at the same time, the novelties of the current research are also presented in this table.

4- The authors of this article have a total of more than 40,000 citations in the field of modeling, environment and agriculture and have done many studies in these fields. Almost all of them were in the range of one and two percent of the world's top highly-cited scientists on September 2022 (based on Stanford University Report). It is obvious that unconsciously a number of their studies will be present in a comprehensive literature review. Moreover, I am an editor at Elsevier and I am very surprised by Subject Editor of Agronomy because according to the principles of peer-review journals, there should not be send the manuscripts for reviewers with author’s names because personal relations may also influence the results of their reviewing. Of course, fortunately, you were an exception to this issue.

Best Regards

Author Response File: Author Response.docx