Use of Plackett–Burman and Box–Behnken Designs to Optimize Bioelectricity Production from Winery Residues
Round 1
Reviewer 1 Report
The paper is interesting and the authors performed a good work to (Use of Plackett-Burman and Box-Behnken designs to optimize bioelectricity production from winery residues) in Water journal-2023, there are a few comments need to be addressed as follows:
1. The abstract needs revision and some results.
2. Add goodness-of-fit of the models.
3. In the introduction sentence from lines 49-86. Its place here is inappropriate. I suggest moving it to the Method section
4. Table 1 is repeated. I suggest that you delete it from the introduction.
5. Introduction expresses, explains, and compares a priori studies and the current study to justify the research's need.
6. Confirmed from superscript and sub-script in the equations and standard units.
7. It is recommended to review references according to the journal's editorial guidelines ( as line 85).
8. Put suitable references in 2.2 & 2.3 for both Placket-Burmann and Box-Behnken factorial designs.
9. Analysis of variance (ANOVA) for a selected factorial model for the Placket-Burman design should be presented.
10. In the results section the data should be presented as concisely as possible as R2 value, P - Value, and show validation of the model.
11. The discussion section should give an interpretation of the significance of the results obtained with reference to similar works done by other authors.
12. Figures 5-7 need to improve.
Moderate editing of English language required.
Author Response
Review report 1: comments to revisions
The paper is interesting and the authors performed a good work to (Use of Plackett-Burman and Box-Behnken designs to optimize bioelectricity production from winery residues) in Water journal-2023, there are a few comments need to be addressed as follows:
Answer: Thank you very much for your comments. We hope the manuscript will contribute to increase knowledge about the use of winery residues as sources of energy, which add a new revalorisation technique to these materials.
We would like to thank the comments made by the reviewer, which will help us to improve the overall quality of the manuscript and that we have considered in this review. In order to facilitate review, we would like to indicate you that changes in the manuscript after your suggestions have been highlighted in green (other colours correspond to the other reviewers).
Specific comments and answers:
- The abstract needs revision and some results.
Answer: The abstract has been revised to include the main results, variables analysed and important values (page 1). We believe that it is now more informative and contains the purpose, methodology and main results of our experiments.
- Add goodness-of-fit of the models.
Answer: Following reviewer suggestions, p-significance levels have been added to the text, both for Plackett-Burman (page 8) and Box-Behnken Design (pages 11) designs. Consequently, two new tables have been added to the text (Tables 4 and 6) which include the regression coefficients and their statistical significance. Significant coefficients were set at a p<0.05 level and marked with asterisks.
- In the introduction sentence from lines 49-86. Its place here is inappropriate. I suggest moving it to the Method section.
Answer: Following reviewer suggestion, introduction has been rewritten and those parts describing BBD and PB have been moved to the methods section (pages 4-5). Also, as suggested in comment #5, introduction has been changed and focused mainly on the bioelectricity production.
- Table 1 is repeated. I suggest that you delete it from the introduction.
Answer: Table 1 has been deleted and numbering of the rest of the tables revised.
- Introduction expresses, explains, and compares a priori studies and the current study to justify the research's need.
Answer: Following reviewer suggestions, details about Plackett-Burman and Box-Benhken designs have been moved to the methodology section. Consequently, introduction has been increased by including the need to find solutions for wine residues. New information about the toxicity of residues and the need to find new solutions for these residues has been added to the text (page 1).
- Confirmed from superscript and sub-script in the equations and standard units.
Answer: The document has been revised and now there are no missing super- or subscripts.
- It is recommended to review references according to the journal's editorial guidelines ( as line 85).
Answer: References have been revised and formatted according to the journal´s style. Also, some new references have been added.
- Put suitable references in 2.2 & 2.3 for both Placket-Burmann and Box-Behnken factorial designs.
Answer: Correct references for PB and BBD have been added to the text (page 4, line 159; page 5, line 190) and to the references section (page 14, references number 17 and 19).
- Analysis of variance (ANOVA) for a selected factorial model for the Placket-Burman design should be presented.
Answer: Following reviewer suggestions, ANOVA analyses for PB have been included in the results section. In general, results have been re-organized to follow your suggestions and regressions coefficients, p-values and validation graphs of the model are included after the table results. This sequence has been followed both for PB and BBD.
- In the results section the data should be presented as concisely as possible as R2 value, P - Value, and show validation of the model.
Answer: Following the reviewer suggestions, the results section has been reorganized: for each model (PB and BBD) now the following sequence is presented: first, results of the analyses; second, Pareto chart of effects showing the most significant factors; (3) validation graphs of the model showing the predicted vs. observed values of the model; (4) p-values showing the significant factors.
- The discussion section should give an interpretation of the significance of the results obtained with reference to similar works done by other authors.
Answer: We have included more discussion in this section. It is discussed about the findings of other studies related to the variables of our study: we discuss about the nature of vinasse and options other authors have carried out to solve the sedimentation of the vinasse, such as adding polymers or porous solids (5 references in this paragraph). Also, the effect of temperature, which seemed to affect negatively to bioelectricity production has also been included in this section, and there is a consensus on the need to employ room temperature, to avoid excessive microorganisms ‘growth (4 references). Our findings about the effect of salt, which showed little effect on the system, suggest that the use of this MFC could be expanded to saline waters without affecting its yield, although references included in this paragraph (1 reference) suggest researching more in this field. Finally, other authors have included yeast as an important factor in the MFC yield, suggesting that yeast present a higher ability to grow in harsh environments than most bacteria making these systems more robust (1 reference). However, in our case, vinasse is rich in yeast and bacteria which was unaffected by the natural functioning of the MFC, so the addition of external yeast seemed to be negligible. Also, the importance of the use of MFC in the literature has also been highlighted (7 references). The manuscript cites 42 documents in total, although if you miss some discussion, please don´t hesitate to let us know, in case there is a second revision.
- Figures 5-7 need to improve.
Answer: Following the reviewer comments, figures have been significantly improved and now they are easier to read and interpret. Also, following comments from reviewer #2 the figure 9 has been simplified.
Reviewer 2 Report
Introduction
idea in line 34 must be rewritten
please change the word "deduce statistics"
please change "study the interaction between" for "identify the interactions between"
eliminate "So" in line 39
please rewrite the sentence "to 40 achieve the best efficiency of a system to obtain the best possible result."
remove "optimization" from line 57, PB designs are not optimization designs, but screening desings
rewrite "which aims at determining the influence of 57 certain variables in an experiment"
the paragraph in lines 68-73 must be removed from the introduction, since it does not add critical information to the introduction, but it does describe the statistical method
the same applies to paragraph 87-107
Methods
there is a repetitive mistake in the methods, since authors mixed "stirring" and "agitation", please choose only one in both figures and text
which version of STATISTICS?
How many center points were used in the B-B design?
RESULTS
when working with DoE, both Pareto and ANOVA data are important to determine which variables are the most critical, in the case of figure 5 the ANOVA data is required to determine which variables are the most significant
if only the data at 15 min are used for the model, the data from 1-10 min are irrelevant, please remove them
in the B-B analysis, in figure 6d there are three agitation marks, and it should be only 2 (Lineal and Quadratic), therefore in post-edition there was a mistake. also please add the ANOVA chart to identify the P value of agitation and concentration variables
figure 7 must be re-done, since only agitation and concentration are the most significant variables, therefore the other interactions can be removed
the paper has not been optimized, since the curvature is far from the center, therefore a new B-B design (better a CCD with 5 central points) must be done in order to really identify the actual optimization area
CONCLUSIONS
this section is too large, please simplify and highlight the main results only
Several grammatical errors must be overcome, this is a list of some of them:
Introduction
idea in line 34 must be rewritten
please change the word "deduce statistics"
please change "study the interaction between" for "identify the interactions between"
eliminate "So" in line 39
please rewrite the sentence "to 40 achieve the best efficiency of a system to obtain the best possible result."
remove "optimization" from line 57, PB designs are not optimization designs, but screening desings
rewrite "which aims at determining the influence of 57 certain variables in an experiment"
Author Response
Review report 2:
First of all, we would like to thank the comments made by the reviewer, which will help us to improve the overall quality of the manuscript and that we have taken into account in this review. In order to facilitate review, we would like to indicate you that changes in the manuscript after your suggestions have been highlighted in yellow (other colours correspond to the other reviewers).
Specific comments:
INTRODUCTION
- Idea in line 34 must be rewritten
Answer: First lines have been rewritten to explain the need to employ optimization techniques. We keep the reference [1] because it explains the benefits of experimental designs and it is still useful.
- Please change the word "deduce statistics"
Answer: The above term has been changed to “obtain statistical parameters” (page 1, line 41).
- Please change "study the interaction between" for "identify the interactions between"
Answer: The suggestion has been carried out (page 1, line 41).
- Eliminate "So" in line 39
Answer: The suggestion has been carried out (page 1, line 43).
- Please rewrite the sentence "to achieve the best efficiency of a system to obtain the best possible result."
Answer: The sentence has been rewritten as follows: “Optimization techniques arises as a set of methods to improve the operation of systems and to identify the optimum values of variables influencing its performance” (page 1, lines 43-44).
- Remove "optimization" from line 57, PB designs are not optimization designs, but screening designs
Answer: The suggestion has been carried out (page 2, line 59).
- Rewrite "which aims at determining the influence of certain variables in an experiment"
Answer: The suggestion has been carried out (page 4, section 2.2). Also, according to the comments of reviewer #1 description of Plackett-Burman and Box-Behnken design has been moved to the methodology section.
- The paragraph in lines 68-73 must be removed from the introduction, since it does not add critical information to the introduction, but it does describe the statistical method. The same applies to paragraph 87-107
Answer: Description of Plackett-Burman and Box-Behnken Designs has been moved to the methodology section (pages4-5, sections 2.2 and 2.3).
METHODS
- There is a repetitive mistake in the methods, since authors mixed "stirring" and "agitation", please choose only one in both figures and text
Answer: Thank you very much for your comment: we have revised the manuscript and changed “agitation” to “stirring” in the text and figures.
- Which version of STATISTICS?
Answer: The correct version of the software employed has been included in the text (page 5, lines 179).
- How many center points were used in the B-B design?
Answer: In this study an incomplete 33 factorial design was carried out with 12 combinations of the three independent variables and three replicates of one central point. This information has been added to the text in page 5, lines 205-207.
RESULTS
- When working with DoE, both Pareto and ANOVA data are important to determine which variables are the most critical, in the case of figure 5 the ANOVA data is required to determine which variables are the most significant.
Answer: Thank you very much for your suggestion. ANOVA data has been included in the manuscript in tables 4 and 6 and the text has been rewritten to include comments about these tables. Also, validation graphs of the model have been included, as suggested by reviewer #1.
- If only the data at 15 min are used for the model, the data from 1-10 min are irrelevant, please remove them.
Answer: We have included data for 1 to 15 min and optimization graphs are also for the four times analysed. We believe that it is important to follow the progress with time to control instant production and also to measure the fluctuations in bioelectricity production. A little explanation about this need is included in page 6, lines 223-228.
- In the B-B analysis, in figure 6d there are three agitation marks, and it should be only 2 (Lineal and Quadratic), therefore in post-edition there was a mistake. also please add the ANOVA chart to identify the P value of agitation and concentration variables
Answer: Thank you very much for your comment: we have edited the labels, as the software exports the graphs in bad quality, and there has been a mistake in one of them. This mistake has been corrected and also ANOVA data has been included in tables 4 and 6.
- Figure 7 must be re-done, since only agitation and concentration are the most significant variables, therefore the other interactions can be removed
Answer: Figures related to surface responses of 1 and 5 min have been removed and discussion of text will also have been corrected (to avoid references to the deleted figures) (page 12, figure 9).
- The paper has not been optimized, since the curvature is far from the center, therefore a new B-B design (better a CCD with 5 central points) must be done in order to really identify the actual optimization area
Answer: We agree with your comments related to the central point of the curvature. Maybe we have employed some expressions leading to confusion.
We want to evaluate the behaviour of a system within the first 15 min of experiment. On the one hand, references in bibliography relate electricity conditioned of a series of factors and we wanted to evaluate, in the first part, which of them were significative. On the other hand, we wanted to determine the instantaneous bioelectricity production in the first minute, followed by measurements at different intervals, up to 15 min. We found out that, in the first minute, just as the beginning of the experiment, vinasse concentration positively influences the experiments but, as it is a complex system, its suspended material tends to settle with time, and this is the reason that stirring becomes more important with time.
So, at his time of our experimentation, what we really want to research is the influence of the variables on the bioelectricity production at each time evaluated and our focus is to maximize bioelectricity in the context of the variables analysed. So, we agree with the fact that “optimize” may lead to confusion as we research was conducted to research which variables increment millivolt production (and also which combination produces a significant decrease).
Although we really appreciate your comment about optimization, we think that full optimization could be part of a future research because we are involved now in a series of assays to evaluate the long-term behaviour of the MFC with a system manufactured by ourselves (with 3D-additive manufacturing).
So, this focus will be corrected in the manuscript, avoiding the term “optimization of the process” to express that we want to maximize the response of the system.
CONCLUSIONS
- This section is too large, please simplify and highlight the main results only
Answer: Conclusions have been rewritten, emphasizing the main results of the manuscript (pages 13, conclusions section).
COMMENTS ON THE QUALITY OF ENGLISH LANGUAGE
Several grammatical errors must be overcome, this is a list of some of them:
INTRODUCTION
- Idea in line 34 must be rewritten
Answer: First lines have been rewritten to explain the need to employ optimization techniques. We keep the reference [1] because it explains the benefits of experimental designs and it is still useful.
- Please change the word "deduce statistics"
Answer: The above term has been changed to “obtain statistical parameters” (page 1, line 41).
- Please change "study the interaction between" for "identify the interactions between"
Answer: The suggestion has been carried out (page 1, line 41).
- Eliminate "So" in line 39
Answer: The suggestion has been carried out (page 1, line 43).
- Please rewrite the sentence "to achieve the best efficiency of a system to obtain the best possible result."
Answer: The sentence has been rewritten as follows: “Optimization techniques arises as a set of methods to improve the operation of systems and to identify the optimum values of variables influencing its performance” (page 1, lines 43-44).
- Remove "optimization" from line 57, PB designs are not optimization designs, but screening designs
Answer: The suggestion has been carried out (page 2, line 59).
- Rewrite "which aims at determining the influence of certain variables in an experiment"
Answer: The suggestion has been carried out (page 4, section 2.2). Also, according to the comments of reviewer #1 description of Plackett-Burman and Box-Behnken design has been moved to the methodology section.
Reviewer 3 Report
The table at pag 8/14 is numbered as 2, I think there is a mistake. It may be Table 5.
The values reported in tables 4 and 5 (if previous comment is right) era mean value of how much measures?
Please, add acronyms and symbols list.
Author Response
Review report 3:
First of all the authors would like to thank the comments made by the reviewer, which will help us to improve the overall quality of the manuscript and that we have taken into account in this review. In order to facilitate review we would like to indicate you that changes in the manuscript after your suggestions have been highlighted in blue (other colours correspond to the other reviewers).
Specific comments:
- The table at pag 8/14 is numbered as 2, I think there is a mistake. It may be Table 5.
Answer: Thank you very much for your comment: we have revised reference numbers of both tables and figures and now they are all correct.
2- The values reported in tables 4 and 5 (if previous comment is right) era mean value of how much measures?
Answer: Experiments were run in triplicates, as we expected that results would present high variability. Also, Arduino recorded values each 10 seconds (from starting time to 15 minutes) and we employed three measurements for each time interval, e.g., for time 1 minute we employed the 50 s, 1 min and 1 min 10 s. In tables 4 and 5 (now tables 3 and 5) we present the average value ± the coefficient of variation. So, this information has been added to the methodology section in page 5.
3- Please, add acronyms and symbols list.
Answer: In relation with this comment, we understand the difficult to the reader to follow the text when PB and BBD are continuously employed, which are acronyms not very common or familiar to the reader. So, both acronyms were full named throughout the text and their acronyms have been deleted. Also, DoE only was employed twice in the manuscript and has also been replaced by “Design of Experiment”. The other acronym which has been consistently employed is MFC (Microbial Fuel Cell) although this is very common and familiar to the reader. Moreover, following the journals’ style we have placed the full name and the acronym in each section. So, we apologize for not placing a table with the acronyms and symbols list at the beginning of the text, but we have corrected the text to make it more understandable and, also, we didn´t want to increase the length of the manuscript which has actually 9 figures and 6 tables.
Round 2
Reviewer 1 Report
There is no objection to publishing (Accept in present form), the authors have modified the required questions.
Minor editing of English language required.
Reviewer 2 Report
Authors have significantly improved the quality and presentation of the document
