ANN Hybrid Model for Forecasting Landfill Waste Potential in Lithuania
Round 1
Reviewer 1 Report
The abstract is poorly structured and does not give a proper idea of the research being presented. Please follow the general pattern of abstract composition: [problem definition] -> [concept] -> [action] -> [results] -> [significance] - in smooth text.
Please avoid using acronyms in the abstract, except well-known ones as ANN.
There are reference lumps in the text such as: [2, 3, 4]. Please eliminate this lump. After that please check the manuscript thoroughly and eliminate ALL the lumps in the manuscript. This should be done by characterising each reference individually. This can be done by mentioning 1 or 2 phrases per reference to show how it is different from the others and why it deserves mentioning. This is not just a formalism. Having reference lumps in the text casts a serious doubt of whether the authors have really read and understood the cited sources. If you do not characterise the references individually it matters little how they are formatted. What really matters is to have meaningful references and the requirement for individual characterisation aims exactly at that. Even the small lumps leave something unsaid and reduce the quality and impact of the paper. Note - just adding the author names is not a sufficient distinguishing characterisation of the references.
In the introduction section, the relevance reasoning for the presented research is fine. However, the paper goal formulation is quite vague - just mentioning that a novel forecasting solution is presented. However, the novelty reasoning itself is missing. What is the novel contribution? What existing knowledge gaps have been identified? How are they proposed to be overcome and state of the art progress to be achieved?
The figures use an excessively tiny font, I am struggling to see and comprehend their content.
Before proceeding to describe your chosen model and actions, please describe your scientific hypothesis, concepts and the relevant reasoning for choosing the particular modelling approach. This should be accompanied by an overall description of the followed procedure. A block diagram of the procedure would be also very useful.
The method description reads with difficulty and I cannot see a systematic model. There is a figure illustrating part of the forecasting logic (Figure 6). But beyond that, the rest is unclear and cannot be followed.
Please minimise the use of acronyms and always define them at the point of their first appearance.
What are the grounds for formulating Equation 3? Please provide proper references or a proper derivation. Overall, please present the method properly and with appropriate references and derivations. The current version is sporadic at best.
What results is the "Results" section supposed to present? Please adapt this section to clearly illustrate your proposed method after providing the method presentation in a correct and comprehensible way.
Author Response
We appreciate the time and effort on the part of the editor and referees in reviewing this manuscript and providing constructive comments that helped to improve and clarify the manuscript considerably. We hope that the revised version of the manuscript answers their concerns.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Title: Forecasting Model of Energy Recovery Potential from Landfill 2 Waste: The Lithuanian Case Study
Manuscript ID: sustainability-1614055
Type: case report
General comment:
The paper reports an ANN modeling study on waste energy content based on waste composition using disposal records from 2007 to 2020 at Lapes landfill and Aukstrakiai landfill in Lithuania. The concept of waste-to-energy is important in sustainable waste management. The methods are reasonable. There are some minor language and formatting issues, but the manuscript is generally well written. The major weaknesses are (i) lack of novelty, (ii) low reliability of the modeling results, and (iii) ineffective use of tables and figures.
- Lack of novelty. The method appears reasonable, but they are very similar to other ANN waste studies. The reviewer is not able to recognize the novelty of the work. A comprehensive literature review is missing, and the potential knowledge gap is not identified. A concise novelty statement must be provided in the text.
- The key results are not consistent with everyday observations. For example, the paper estimated a significant energy peak in 2015 using an empirical equation (Figure 10). The 2015 peak is about 800 times higher than 2014, and about 8 times higher in 2017. This is highly unlikely. I suggest the authors to check the quality of input data. In addition, Figure 11 clearly shows that the local optimums are reported.
- Many figures are too general and add little to the study objective. Discussion is often general. Adding too much non-essential materials only dilute the focus and impact of your study. Please see specific comments below.
I believe more work is needed.
Specific comments:
L51. “Several studies have been conducted to develop a forecasting model based on mentioned indicators…” The literature review is very weak. Please check recent waste modeling studies using RNN, LSTM, ANN published in the last 2 years in the following journals: Waste Management, Sustainable Cities and Society, Science of The Total Environment.
L60. The results shown in Figure 2 is problematic. Please provide the methodological details. For examples, how many published studies are considered? From which periods? How the studies were selected? Please remove the Figure, or add methodological details.
L72. Figures 2 and 3 are misleading. I suggest you to use the waste amount per capita (kg/cap) in your y-axis.
L73. In Figure 2, why there was a significant drop in waste disposed in 2016? The number was over 1 million tonnes in 2011, and dropped to 0.38 million tonnes in 2016. Do you have a significant drop in population within 5 years?
L109. The units of the socio-economical factors (Employment, Emigration, Immigration, etc.) in Table 1 must be provided. Are you using the actual numbers, or percentages?
L110. The acronym “R&D” in Table 1 should be defined.
L139. “The recorded amount of each fraction expressed in tons…” Tons or Tonnes?
L132, 143, 150. The Figures 4, 5, 6 are general and add little to your study. You may want to remove, or combine them to improve the flow of the paper.
L199. The calculation of the energy content is confusing. A total of 11 fractions are shown in Table 2. However, only 7 of them are considered in Equation 4.
L210-221. This is clearly methodology. Please remove it from the Results section.
L217. The training:validation:testing must be provided. Has cross validation test conducted?
L257. The use of lagged inputs in ANN models has been reported in literature and should be discussed. Check literature and compare results with other published studies.
L280. The energy estimates are questionable. Your 2015 peak is about 800 times higher than 2014, and about 8 times higher in 2017. Please check input quality.
L289. “The best forecasting results were achieved using models with 6 inputs.” Have you considered using only 4 or 5 inputs? How do you know they are “the best”?
L307. As shown in Figure 11, local optimums are reported, and the results are misleading. The reviewer understand that may be its too late to make the adjustment. At a minimum, a detailed limitation section should be added to the manuscipt.
Author Response
We appreciate the time and effort on the part of the editor and referees in reviewing this manuscript and providing constructive comments that helped to improve and clarify the manuscript considerably. We hope that the revised version of the manuscript answers their concerns.
Author Response File: Author Response.pdf
Reviewer 3 Report
As explained in this paper, the wastes in landfill have a negative impact on quality of human life. There are many difficulties in effectively treating of these wastes. As such, approaching the field of energy recovery to solve these problems in this paper can be an important solution. It contains a lot of information with the model for forecasting the energy recovery potential. And it will be a sufficient impact in the recycling field of wastes.
Nevertheless, there is some major and minor questions.
Major
1) As the title, the main topic of the paper is forecasting the energy recovery potential. The results and discussions are lacking. Of course, it was explained in Figure 10 of section 3.2 and Table 5 of section 4.
2) The usability of the model is questionable. It's complicated, but I doubt it works well.
3) It is not clear about the meaning and significance of "input" in all data.
4) Explanation of all figure, table and equation are not comfortable.
paper.
Additionally I give miner comments as follows:
- Is Figure 4 a concept of only one type of waste among MSW?
- Figure 4 shows 9 inputs (with GDP, population, and amount). Is it the same as the inputs shown in Table 3, Table 4, etc.?
- Does Figure 5 represent the entire MSW?
- What are the abbreviations of MSE, RMSE, and MAE?
- Missing "N" and "W" in 186-187pp.
- Missing "S" in 194-196pp.
- The contents of 199-208pp overlap with Figure 6.
- The contents of 210-221pp should be dealt with in more detail in "2. Materials and Methods"
- In Tables 3 and 4, does “Input” mean the factor of Figure 4? Or does it mean month?
- What is the difference between time[months] in Figure 7 and 8 and lag[month] in Figure 9?
- Why are 20.6 MJ/kg and 8.5 MJ/kg in 268-269pp different from the values in Table 2?
- In 276pp, Figure 9 à Figure 10
- In all table and figure, terms of factor, conditions, etc. need to be explained.
- In Table 4, are the data from Aukstrakiai? Is it from Lapes? or combined? And what does "5 different forecasting models" mean?
- It is difficult to understand the sentences of 305-306pp. What is the difference between fixed and flexible?
Author Response
We appreciate the time and effort on the part of the editor and referees in reviewing this manuscript and providing constructive comments that helped to improve and clarify the manuscript considerably. We hope that the revised version of the manuscript answers their concerns.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Method and abstract descriptions are better, thank you.
Please use SI units and SI-compliant symbols for widely used non-SI units. E.g. "tonnes" symbol should be "t" and the litres "L", “minutes” –> “min”, years -> “y”, hours -> h, etc. The dimensionless symbol is "(1)". Make sure to leave a space between every value and its symbol for the measurement unit.
Please correct the spelling of your text - there are typos.
I am sorry - I still cannot see a justification of the novelty of the current work, or a clear formulation of the goal for the current work.
Author Response
Thank very much for your comments and suggestions. We fixed our paper accordantly.
Reviewer 2 Report
Manuscript title: ANN Hybrid Model for Forecasting Landfill Waste Potential: The Lithuanian Case Study
Manuscript ID: sustainability-1614055
Type: Case Report
General comment:
The authors have addressed most of the Reviewer 2’s questions in the first round of review. Most responses given are direct and specific. Manuscript has also been revised.
The paper is organized and generally well-written. The major weakness is the novelty, which is difficult to address at this stage of publication. Although the methods are well established, but the numerical results are original. For a case report, I believe this is acceptable. The work also has good practical implications. I believe the work is of interest to the readers of this journal.
I suggest acceptance.
Author Response
Thank very much for your comments and suggestions how to improved an article.
Reviewer 3 Report
The question presented in the first review was well answered and it seems that sufficient revisions have been made.
And English language and style are fine/minor spell check required.
Author Response
Thank very much for your comments and suggestions how to improved an article.
