Empirical Safety Stock Estimation Using GARCH Model, Historical Simulation, and Extreme Value Theory: A Comparative Study
Round 1
Reviewer 1 Report (Previous Reviewer 1)
This version is improved from the previous one; while I still have some small issues with the way that the results are presented I think it might be a more stylistic thing with respect to how my field presents these types of results and their significance than anything incorrect on the part of the authors.
I'm still not convinced of the novelty either, but given the context of the cited papers I'm inclined to believe that there is sufficient evidence of novelty to justify publication after a heavy edit.
Author Response
Journal: Applied Sciences
Manuscript Status: Pending major revisions
Manuscript ID: applsci-1903539
Title: Empirical Safety Stock Estimation using GARCH model, Historical Simulation, and Extreme Value Theory
We thank the editor and the four reviewers for his/her overall support of the study and for the insightful comments and suggestions provided. The novel version takes into consideration all the comments provided. Please see below our response to the comments, and modifications in ref color in the text.
Author Response File: 
Author Response.pdf
Reviewer 2 Report (New Reviewer)
The document is interesting, but at the same time extensive. I could consider sending a lot of data to the annexes. And be specific at the beginning, to better focus on the item of results and finals.
Improve the quality of some images.
Author Response
Journal: Applied Sciences
Manuscript Status: Pending major revisions
Manuscript ID: applsci-1903539
Title: Empirical Safety Stock Estimation using GARCH model, Historical Simulation, and Extreme Value Theory
We thank the editor and the four reviewers for his/her overall support of the study and for the insightful comments and suggestions provided. The novel version takes into consideration all the comments provided. Please see below our response to the comments, and modifications in ref color in the text.
Author Response File: Author Response.pdf
Reviewer 3 Report (New Reviewer)
The article and the research carried out seem very interesting, although I am not an expert in the field of modeling. However, I also conduct research on the use of the simulation method in decision making. So I read it with great interest.
Comments
„Customer service is frequently used in literature” – did the authors say that „The level of customer service measured by the availability of stocks is frequently used in literature. What you mean is that normally the level of safety stock is calculated by which a certain level of customer service is zapewniony? And an alternative is calculating total costs which include the so called „lost sales” resulting from poor service? But it is diffult to asses this costs?
You wrote:
„This work proposes to determinate safety stock using two methods mentioned above, which are also largely applied in finance and insurance domains to estimate Value at Risk which is defined as the quantile of the studied distribution for given confidence level.”
So, your contribution is that, taking into account the limitations and shortcomings of the methods used so far to optimize inventories, you have proposed to solve this problem methods that are already developed, but in a different area - i.e. in the area of finance? However, Trapero has already used it, but you propose a modification?
Therefore, the final objective of this research can be summarized by the following primary research question: How can we improve the safety stock estimation to simultaneously consider the heteroskedasticity phenomena and the occurrence of extreme demands?
So so far no one has yet taken into account: „the heteroskedasticity phenomena and the occurrence of extreme demands”?
Equation (4) of the safety stock is the most useful in the literature, although Equation (5) offers an accurate relationship. One of the major drawbacks of the theoretical approach is that it disregards the fact that the forecast errors are not i.i.d. in reality. In addition, the true model of each item is not found and also the choice of the forecasting error model cannot be under the control of the firm. To overcome the i.i.d. problem of the forecast errors and the problem when the distribution of forecast errors is unknown, a second empirical approach is considered in which neither the point forecast model nor its parameters must be known. This is mainly beneficial in practice, especially when such information is not provided to users
Since in the surveyed company the real demand is not normally distributed, what is the distribution of demand?
I would also like to ask the authors, because I am looking for information on the characteristics of the demand myself. Are there studies on the probability distribution of demand for a given item, perhaps broken down by industry? Does Gaussian and Gamma distribution actually occur in practice?
Author Response
Journal: Applied Sciences
Manuscript Status: Pending major revisions
Manuscript ID: applsci-1903539
Title: Empirical Safety Stock Estimation using GARCH model, Historical Simulation, and Extreme Value Theory
We thank the editor and the four reviewers for his/her overall support of the study and for the insightful comments and suggestions provided. The novel version takes into consideration all the comments provided. Please see below our response to the comments, and modifications in ref color in the text.
Author Response File: Author Response.pdf
Reviewer 4 Report (New Reviewer)
- The abstract needs to be improved. The first sentence in the abstract, it is necessary for the authors to add a sentence to describe the problem or motivation to focus on this topic. The second sentence should provide the literature gap. In the third sentence, the authors should say what you are doing, and then provide the empirical findings. Finally, the significance of the finding should be offered.
- There is no flow in the text. It partly depends on the lack of proofreading but also on the fact that many statements and claims are made without being followed up by a clear and logical discussion. It is especially problematic in the Introduction that brings up a number of findings from different areas without linking them together.
- In the introduction, you need to connect the state of the art to your paper goals. Please follow the literature review by a clear and concise state of the art analysis. This should clearly show the knowledge gaps identified and link them to your paper goals. Please reason both the novelty and the relevance of your paper goals. Clearly discuss what the previous studies that you are referring to. What are the Research Gaps/Contributions? Please note that the paper may not be considered further without a clear research gap and novelty of the study.
- Literature Review has the chance to be further improved: it seems that the authors have made the retrospection. However, via the review, what issues should be addressed? What is the current specific knowledge gap? What implication can be referred to? The above questions should be answered. Authors need to propose their study. You can use: Sustainable Portfolio Optimization Model Using PROMETHEE Ranking: A Case Study of Palm Oil Buyer Companies. Discrete Dynamics in Nature and Society, 2022. Designing a sustainable closed-loop supply chain network considering lateral resupply and backup suppliers using fuzzy inference system. Environment, Development and Sustainability, 1-34. Designing an integrated responsive-green-cold vaccine supply chain network using Internet-of-Things: artificial intelligence-based solutions. Annals of Operations Research.
- Please make sure your conclusions' section underscore the scientific value added of your paper, and/or the applicability of your findings/results, as indicated previously. Please revise your conclusion part into more details. Basically, you should enhance your contributions, limitations, underscore the scientific value added of your paper, and/or the applicability of your findings/results and future study in this session. The discussion is relatively simple and insufficient. I recommend strengthening the comparison with previous research. Please compare the results in this study with those in previous studies. Discuss the study findings here. The discussion and conclusion are appropriately written and require no changes. The manuscript does not answer the following concerns: Why is it timeliness to explore such a study? What makes this study different from the previously published studies? Are there any similarly findings in line with the previously published studies? Are the findings different from prior academic studies that were conducted elsewhere, if any?
Author Response
Journal: Applied Sciences
Manuscript Status: Pending major revisions
Manuscript ID: applsci-1903539
Title: Empirical Safety Stock Estimation using GARCH model, Historical Simulation, and Extreme Value Theory
We thank the editor and the four reviewers for his/her overall support of the study and for the insightful comments and suggestions provided. The novel version takes into consideration all the comments provided. Please see below our response to the comments, and modifications in ref color in the text.
Author Response File: Author Response.pdf
Round 2
Reviewer 4 Report (New Reviewer)
The abstract needs to be improved. The first sentence in the abstract, it is necessary for the authors to add a sentence to describe the problem or motivation to focus on this topic. The second sentence should provide the literature gap. In the third sentence, the authors should say what you are doing, and then provide the empirical findings. Finally, the significance of the finding should be offered.
- There is no flow in the text. It partly depends on the lack of proofreading but also on the fact that many statements and claims are made without being followed up by a clear and logical discussion. It is especially problematic in the Introduction that brings up a number of findings from different areas without linking them together.
- In the introduction, you need to connect the state of the art to your paper goals. Please follow the literature review by a clear and concise state of the art analysis. This should clearly show the knowledge gaps identified and link them to your paper goals. Please reason both the novelty and the relevance of your paper goals. Clearly discuss what the previous studies that you are referring to.
What are the Research Gaps/Contributions? Please note that the paper may not be considered further without a clear research gap and novelty of the study.
- Literature Review has the chance to be further improved: it seems that the authors have made the retrospection. However, via the review, what issues should be addressed? What is the current specific knowledge gap? What implication can be referred to? The above questions should be answered.
Authors need to propose their study. You can use: Sustainable Portfolio Optimization Model Using PROMETHEE Ranking: A Case Study of Palm Oil Buyer Companies. Discrete Dynamics in Nature and Society, 2022. Designing a sustainable closed-loop supply chain network considering lateral resupply and backup suppliers using fuzzy inference system. Environment, Development and Sustainability, 1-34. Designing an integrated responsive-green-cold vaccine supply chain network using Internet-of-Things: artificial intelligence-based solutions. Annals of Operations Research.
- Please make sure your conclusions' section underscore the scientific value added of your paper, and/or the applicability of your findings/results, as indicated previously. Please revise your conclusion part into more details. Basically, you should enhance your contributions, limitations, underscore the scientific value added of your paper, and/or the applicability of your findings/results and future study in this session. The discussion is relatively simple and insufficient. I recommend strengthening the comparison with previous research. Please compare the results in this study with those in previous studies. Discuss the study findings here. The discussion and conclusion are appropriately written and require no changes. The manuscript does not answer the following concerns: Why is it timeliness to explore such a study? What makes this study different from the previously published studies? Are there any similarly findings in line with the previously published
studies? Are the findings different from prior academic studies that were conducted elsewhere, if any?
Author Response
Journal: Applied Sciences
Manuscript ID: applsci-1903539
Title: Empirical Safety Stock Estimation using GARCH model, Historical Simulation, and Extreme Value Theory
The authors would like to thank the editor and the anonymous reviewers, whose insightful comments and constructive suggestions helped us to significantly improve the quality of this paper. Every change in the text is colored in red
Author Response File: Author Response.pdf
Round 3
Reviewer 4 Report (New Reviewer)
No comment
Author Response
Journal: Applied Sciences
Manuscript ID: applsci-1903539
Title: Empirical Safety Stock Estimation using GARCH model, Historical Simulation, and Extreme Value Theory: A comparative Study
The authors would like to thank the editor and the anonymous reviewers, whose insightful comments and constructive suggestions helped us to significantly improve the quality of this paper.
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
Please see the attached word document.
Comments for author File: Comments.pdf
Reviewer 2 Report
The scientific material presented addresses the important issue of stock optimization and ensuring the continuity of supply chains. An interesting method is proposed, validated by high-performance econometric tests, developed, applied and illustrated by a case study that validates the superiority of the proposed method.
Comments for author File: Comments.pdf
Reviewer 3 Report
The author proposed and verified a new method of calculating safety stock using two models.
What the author does may be a meaningful experiment in some special cases.
However, the author seems to lack an adequate explanation of the theory of inventory control and safety stock.
When calculating safety stock, it should be clearly stated what inventory control policy is used in the warehouse. The method of calculating safety stock differs depending on whether the ordering policy is (r,Q), (s,S) or (T,S).
In the calculation of optimal inventory, the inventory cost calculation formula differs depending on whether the model is a backlog model or an (partial) lost-sales model.
When calculating the safety stock of a supply chain, the formula is different from the Two-echelon safety stock when the bullwhip effect is also considered.
The explanation of lead time is unclear. The calculation method of safety stock differs depending on the relationship between lead time and order interval.
In many inventory management theories, the normal distribution is used by the central limit theorem when demand for multiple periods is combined or when considering errors in demand forecasting. However, in this study, it is assumed that the demand forecast error follows a distribution that is not normal distribution. It is necessary to explain what kind of case this is. It is also necessary to explain what kind of product or service form such a case is. Similarly, when the distribution of errors in the demand forecast is clear, the calculation of the order quantity using the "cycle service level (=percentile)" and the "cumulative distribution function should be written as to why the order quantity should not be calculated using the "cycle service level (=percentile)" and "cumulative distribution function.
Also, the proposed method should be verified that it works correctly even when the error follows a normal distribution.
The theory of safety stock and the theory of the newsvendor problem are difficult to correspond, and in this paper, it is difficult to correctly correspond and evaluate the respective values of profit, loss, critical rate and service rate.
The authors should set a section in the paper the setting of experiment and describe clear description of how the experiment was set and conducted.
