All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
In the opinions of the original reviewers and mine, this revised paper can be accepted.
[# PeerJ Staff Note - this decision was reviewed and approved by Jyotismita Chaki, a PeerJ Section Editor covering this Section #]
It is fine
It is fine
All issues have been resolve
No further comments
The manuscript is well-written and demonstrates significant improvement in language clarity and readability. The background and introduction provide a strong context for the study, effectively outlining the importance of minimizing supply chain costs in e-commerce. The literature review is comprehensive and accurately references recent studies, positioning this research within a broader academic and practical framework.
The experimental design is robust, with a clear, replicable methodology. The authors have improved their description of the methods used, particularly regarding customer segmentation, which now has clear explanations of the autoencoder and random forest models applied. The detailed explanation of the genetic algorithm-enhanced integer linear programming (GA-ILP) model supports transparency and reproducibility.
1. The findings are robust and clearly presented, with conclusions that are directly supported by the results. The use of real-time demand data and the dynamic adjustments in resource allocation are innovative contributions that enhance the relevance and practical applicability of the study.
2. The data is statistically sound and well-controlled, and the framework’s effectiveness is convincingly demonstrated through both model accuracy improvements and reductions in cancellation rates. The limitations are briefly acknowledged, with recommendations for future work, which adds credibility to the findings without overstating their generalizability.
In the opinions of reviewers and mine, a minor revision should be performed on this revised paper.
It is fine
It is fine
The authors stated that the k-fold cross-validation has been applied, however, I could not find the results of k-fold cross-validation in the revised manuscript. Add the results in the manuscript.
no additional comments
The language has improved with clearer and more detailed explanations. The literature review has been enhanced, especially in addressing gaps in prior research. Also, additional figures have been included, such as those illustrating the auto-encoder and random forest model, making it easier for readers to understand the methodology.
The authors have added more specific details about their experimental setup, particularly in the description of the genetic algorithm and integer linear programming (ILP) model. This improves the replicability of the study. And there is now a more comprehensive explanation of how different parameters were optimized, as well as additional examples of how real-time data integration is handled.
1. The revised manuscript includes a stronger discussion of limitations and future work, which helps in framing the study's findings within a broader context.
2. The authors have incorporated additional performance comparisons with other models, making the findings more robust and well-supported.
The revised version has addressed many of these issues, with an expanded literature review, more detailed methodological descriptions, and additional baseline comparisons. The inclusion of more visual aids and examples has also improved the paper's clarity. Minor adjustments could still be made to further enhance clarity and completeness.
In the opinions of revewers and mine, this paper should undertake a major revision to solve all the issues proposed by three reviewers.
1. The abstract needs a few details of how the proposed model works.
2. The Related Work is not described. A separate section focusing on the existing works related to the problem should be discussed. In addition, research gaps should be highlighted clearly.
1. No figure is added for the proposed method.
2. Workflow of the proposed architecture should also be illustrated.
4. The study uses several models including ANN, and CNN along with the proposed framework, however, hyperparameters of these models are need for reproducibility.
5. Details of the dataset are missing like the number of samples, train-test split and number of samples for training and testing.
6. It is not clear whether the curves for accuracy in Figures 4 and 5 are for training samples or testing samples.
7. In Figures, use Epochs instead of Round.
8. Table 4 shows performance of different models but it is not clear "performance for which task"?
1. Performance of the model should be compared with existing approaches from literature.
2. K-fold cross-validation is needed to validate the generalizability and robustness of the proposed approach.
no comments
Clarity and Unambiguity:
The paper contains technical jargon that may not be clear to all readers. Please ensure that all terms are well-defined and that the language is accessible to a broader audience within the field.
Some sections are dense and could benefit from a clearer structure to improve readability. Consider breaking down complex paragraphs into smaller, more digestible parts.
Literature References and Background:
The literature review seems to be somewhat limited. Please expand this section to include a broader range of relevant studies, ensuring that the most recent and pertinent works are cited.
The introduction and background should provide a more comprehensive context for the research. Clearly articulate how your work builds upon and differs from existing literature.
Raw Data Sharing:
We note that the raw data have been shared, which is commendable. However, please ensure that the data are accompanied by descriptive metadata to enhance their utility for future researchers.
Self-Contained with Relevant Results:
The paper should be more explicit in linking the results to the hypotheses or research questions. Ensure that all findings are relevant and that the paper presents a coherent and complete unit of publication.The results section could be strengthened by including more detailed analysis and discussion, possibly with additional experiments or case studies to validate the framework's effectiveness.
Originality and Scope:
Ensure that your research is original and falls within the Aims and Scope of the journal. If there are any aspects of your work that may not be considered original, please clarify how your approach differs from or builds upon existing research.
Research Question:
The research question should be clearly stated in the introduction. It should be relevant to the field and address a meaningful knowledge gap. Please revise the introduction to ensure that the research question is explicitly stated and that its significance is well-articulated.
Knowledge Gap:
The manuscript should clearly identify the knowledge gap that the research aims to fill. Please provide a more detailed discussion of the current state of knowledge in your field and how your work contributes to advancing this knowledge.
Rigor and Ethical Standards:
The investigation must be conducted with rigor and adhere to high technical and ethical standards. Please provide more information on the methodologies used to ensure the ethical conduct of your research, such as data privacy measures or institutional review board approvals.
Methodological Detail:
The methods section should provide sufficient detail for another researcher to replicate the study. Please review this section to ensure that all steps, procedures, and materials are described in detail. If any proprietary methods or materials are used, please discuss how others could access or approximate these for replication purposes.
Impact and Novelty:
The manuscript should clearly articulate the rationale for the research and its potential benefit to the literature. If the study is a replication, please explain why it is necessary and how it adds value. Ensure that the paper does not merely replicate existing work without a clear justification.
Underlying Data:
All data supporting the conclusions must be provided or made available in a suitable repository. Please ensure that the data are robust, statistically sound, and controlled. If there are any limitations to the data, these should be acknowledged and discussed.
Conclusions:
The conclusions drawn in the paper should be directly linked to the original research question and should be supported by the results. Any claims of causation must be backed by well-controlled experimental interventions. Correlations should not be misinterpreted as causation.
Statistical Soundness:
The statistical analysis should be rigorous and appropriate for the data. Please ensure that all statistical tests are correctly applied and that the results are interpreted accurately.
Controls and Replicability:
The experimental design should include appropriate controls to ensure the validity of the results. Please describe the controls in detail and how they ensure the reliability of the findings.
n/a
1. The manuscript is well-structured and clear overall. However, there are occasional instances where the phrasing could be refined for better clarity and readability, particularly in the Introduction and Methodology sections. For example: In the introduction, certain sentences could be rephrased to improve flow and coherence. A thorough proofreading of the manuscript is suggested to enhance clarity.
2. The introduction provides a strong rationale for the study, effectively highlighting the importance of reducing transportation costs in supply chain management. While the literature review is thorough, it would benefit from a more explicit comparison between the proposed ISCCO framework and existing approaches in the field. Specifically, more emphasis could be placed on the unique contributions of this framework. Consider adding a paragraph that clearly differentiates your approach from other methodologies discussed in the literature.
3. The manuscript adheres to the typical structure of an academic paper. Figures are well-designed and relevant, enhancing the reader's understanding of the research. However, Figure 4 could use a more detailed caption to explain the significance of the depicted results. You can improve the figure captions for better self-explanatory clarity.
4. The references are appropriate and up-to-date. Nevertheless, the inclusion of a few more recent studies on deep learning applications in supply chain optimization could strengthen the background section. Consider adding 2-3 recent studies that address similar challenges in supply chain management.
1. The research question is well-defined, and the scope of the study is appropriate. The authors focus on a critical issue in supply chain management and offer a novel solution through the ISCCO framework.
2. The methodology is sound and well-documented. The use of auto-encoders and random forests for customer segmentation is innovative. However, the genetic algorithm’s parameter selection process could be explained in greater detail to enhance the reproducibility of the study. Add a section or subsection that provides a more detailed explanation of the parameter tuning process for the genetic algorithm.
3. The study does not involve human or animal subjects, so there are no major ethical concerns.
1. The data analysis is rigorous, and the results are statistically sound. The authors have demonstrated a significant improvement in user classification accuracy and transportation cost reduction. The use of real-world data strengthens the validity of the findings. However, a discussion on potential limitations, such as scalability issues or applicability to other datasets, would be beneficial. Include a brief discussion on the limitations of the proposed framework, particularly concerning scalability and generalizability.
2. The conclusions are well-supported by the data and are consistent with the research objectives. The findings contribute valuable insights to the field of supply chain management. No major revisions are necessary, but consider reinforcing the discussion on the practical implications of the ISCCO framework in real-world applications.
3. The detailed methodology and availability of underlying data make replication feasible. This is a strong point of the manuscript.
1. The manuscript presents an important contribution to supply chain management, particularly for e-commerce businesses facing challenges in minimizing transportation costs.
2. The integration of deep learning with optimization techniques is commendable, and the results are promising.
3. To further improve the manuscript, the authors might consider discussing potential applications of the ISCCO framework in different industries or scenarios, beyond the presented case study.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.