Recent Trends in the Optimization of Logistics Systems Through Discrete-Event Simulation and Deep Learning
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis paper aims to present the latest trends and research directions in the field of optimization of logistics systems with Discrete-Event Simulation (DES) and Deep Learning (DL).
Please refer to the following comments for improvement.
The main findings should be presented and emphasized in the abstract.
Please provide justification that this review only covers the period from 2015 onwards.
There are grammatical errors found in the paper. Please check and improve the paper.
This study includes both Scopus and Google Scholar databases. However, Web of Science should be included in the analysis due to its importance.
Figures 3-4 is not clear. Kindly improve the visibility of Figures.
A sub section should be added to discuss the evolution of research trends for this topic.
The limitations of this study should be discussed.
The references are appropriate and up-to-date.
Comments on the Quality of English LanguageSome grammatical errors. The English could be improved to more clearly express the research.
Author Response
Comment 1:
The main findings should be presented and emphasized in the abstract.
Reply 1:
Thank you very much for your valuable comment! The abstract was expanded (from lines 21 to 24) to better emphasize the main findings of the study.
Comment 2:
Please provide justification that this review only covers the period from 2015 onwards.
Reply 2:
Thank you very much for your valuable comment! The justification was provided in the introduction from lines 63 to 69, which was further expanded from lines 69 to 72 to provide additional justification.
Comment 3:
There are grammatical errors found in the paper. Please check and improve the paper.
Reply 3:
Thank you very much for your valuable comment! The found grammatical errors were corrected and in general, the English of the paper was improved.
Comment 4:
This study includes both Scopus and Google Scholar databases. However, Web of Science should be included in the analysis due to its importance.
Reply 4:
Thank you very much for your valuable comment! In accordance with your comment, the literature search was expanded with the use of Web of Science, which resulted in the identification and addition of 6 additional papers to the included studies, which now contains 65 articles. The overview of the newly found papers was provided in 4.2, while they were also included in the systematisation and in the diagram of the yearly distribution of the studies.
Comment 5:
Figures 3-4 is not clear. Kindly improve the visibility of Figures.
Reply 5:
Thank you very much for your valuable comment! The Figures were enlarged to provide better visibility. Unfortunately, they could not be enlarged any further due to size constraints, and cutting out smaller parts from them would reduce their informational content. On the other hand, their resolution is sufficient enough for further enlargement by the reader.
Comment 6:
A sub section should be added to discuss the evolution of research trends for this topic.
Reply 6:
Thank you very much for your valuable comment! In accordance with your comment, the discussion (section 5) was split into two sub-sections, the first dealing with the evolution of the research trends. This first sub-section was significantly expanded starting from line 931 and ranging to line 954.
Comment 7:
The limitations of this study should be discussed.
Reply 7:
Thank you very much for your valuable comment! The discussion of the limitations of the study was added to the conclusions, starting from line 1049 and ranging to line 1058.
Comment 8:
The references are appropriate and up-to-date.
Reply 8:
Thank you very much for your valuable comment!
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors present the latest trends and research directions in the field of optimizing logistics systems using Discrete-Event Simulation (DES) and Deep Learning (DL). The article was well-written and read with great interest. One of the benefits of the article is Section 3: "Analysis of Identified Literature Sources", which includes a substantial number of references. The authors conducted a comprehensive review of relevant literature published between 2015 and 2025. The review was based on a systematic approach designed to identify scientific papers that specifically applied discrete-event simulation and deep learning methods for solving logistics problems, or at least offered a conceptual framework that facilitates the joint application of these techniques. At the same time, a disadvantage of the paper is the lack of a concrete example of the implementation of Discrete-Event Simulation and Deep Learning, as a case study, with a brief explanation of its technical and mathematical aspects. Such an example is not necessarily required to be developed personally by the authors, but its absence prevents readers from understanding the intricacies and benefits of this technology. For instance, this could be an example of a simulation model created in AnyLogic, Arena Simulation (or any other simulation tools) that incorporates Deep Learning, or it could be various discrete-event simulation models that use Deep Learning, along with links to relevant sources. Therefore, the following are recommendations for enhancing the article:
1.The literature review could benefit from reviewing more papers that use simulation tools, such as AnyLogic, Arena Simulation, and other open-source discrete-event simulation tools. In this regard, the following relevant articles could be mentioned and cited:
[1] Makarov V.L., Bakhtizin A.R., Beklaryan G.L., Akopov A.S. (2021) Digital plant: methods of discrete-event modeling and optimization of production characteristics. Business Informatics, vol. 15, no 2, pp. 7–20. https://doi.org/10.17323/2587-814X.2021.2.7.20
[2] Sebastian Lang, Tobias Reggelin, Marcel Müller, Abdulrahman Nahhas,
Open-source discrete-event simulation software for applications in production and logistics: An alternative to commercial tools? Procedia Computer Science, Volume 180, 2021,
Pages 978-987, https://doi.org/10.1016/j.procs.2021.01.349.
2. A disadvantage of the paper is the lack of a concrete example of the implementation of Discrete-Event Simulation and Deep Learning, as a case study, with a brief explanation of its technical and mathematical aspects. Such an example is not necessarily required to be developed personally by the authors, but its absence prevents readers from understanding the intricacies and benefits of this technology. For instance, this could be an example of a simulation model created in AnyLogic, Arena Simulation (or any other simulation tools) that incorporates Deep Learning, or it could be various discrete-event simulation models that use Deep Learning, along with links to relevant sources. Therefore, it would be beneficial to include a brief section or sub-section that provides examples of models and systems that utilize both discrete-event simulation techniques and deep learning, implemented within various simulation tools and frameworks. In this section, you can, for example, provide screenshots of the models or their flowcharts.
Author Response
Comment 1:
The literature review could benefit from reviewing more papers that use simulation tools, such as AnyLogic, Arena Simulation, and other open-source discrete-event simulation tools. In this regard, the following relevant articles could be mentioned and cited:
[1] Makarov V.L., Bakhtizin A.R., Beklaryan G.L., Akopov A.S. (2021) Digital plant: methods of discrete-event modeling and optimization of production characteristics. Business Informatics, vol. 15, no 2, pp. 7–20. https://doi.org/10.17323/2587-814X.2021.2.7.20
[2] Sebastian Lang, Tobias Reggelin, Marcel Müller, Abdulrahman Nahhas,
Open-source discrete-event simulation software for applications in production and logistics: An alternative to commercial tools? Procedia Computer Science, Volume 180, 2021,
Pages 978-987, https://doi.org/10.1016/j.procs.2021.01.349.
Reply 1:
Thank you very much for your valuable comment! In accordance with your recommendation, a new section was added to the paper after the introduction titled "Advantages of the combination of Discrete-Event Simulation and Deep Learning in logistics" (starting from line 83), the first sub-section of which deals with the significance of Discrete-Event Simulation in logistics in general, while also mentioning the typically used simulation tools.
Thank you for the recommendation of additional sources. After their analysis, in the mentioned sub-section both of the recommended papers are mentioned and cited as valuable sources for highlighting the importance of DES in relation to logistics.
Comment 2:
A disadvantage of the paper is the lack of a concrete example of the implementation of Discrete-Event Simulation and Deep Learning, as a case study, with a brief explanation of its technical and mathematical aspects. Such an example is not necessarily required to be developed personally by the authors, but its absence prevents readers from understanding the intricacies and benefits of this technology. For instance, this could be an example of a simulation model created in AnyLogic, Arena Simulation (or any other simulation tools) that incorporates Deep Learning, or it could be various discrete-event simulation models that use Deep Learning, along with links to relevant sources. Therefore, it would be beneficial to include a brief section or sub-section that provides examples of models and systems that utilize both discrete-event simulation techniques and deep learning, implemented within various simulation tools and frameworks. In this section, you can, for example, provide screenshots of the models or their flowcharts.
Reply 2:
Thank you very much for your valuable comment! Indeed, in its original form, the paper might have lacked an adequately detailed description of the benefits of the technology. For this reason, a new section was added to the paper after the introduction titled "Advantages of the combination of Discrete-Event Simulation and Deep Learning in logistics" (starting from line 83), the second sub-section of which (starting from line 121) describes in detail the most important application possibilites resulting from the combination of Discrete-Event Simulation and Deep Learning, together with their advantages, benefits and the main characteristics of these applications.
Indeed, a concrete example for the combined application of Discrete-Event Simulation and Deep Learning would aid even more in the understanding of the intricacies and benefits of the specific method which it utilises. Yet, after very careful consideration the description of such an example was eventually not implemented due to the following considerations:
- While a proper description of a concrete example would certainly aid even more in the understanding of the mentioned aspects of a given method, in the view of the author, the adequate introduction of a sufficiently advanced case study would go far beyond the limits of a brief section or sub-section in both length and scope. On the other hand, the paper was submitted to the journal as a "Review" paper, with the specific goal of providing an overview and a systematisation of the related literature and the research trends. The significant expansion of the paper with a sufficiently advanced original case study might not only shift the focus away from the original goals of the study, but it would also make the paper in many ways more similar to a Research Paper.
- Indeed, it would be possible to provide examples from already existing case studies. However, this would on one hand require obtaining consents from multiple different authors, which would obivously significantly complicate the publication of the current study. On the other hand, relying on other studies to such a degree might also negatively impact the originality of the current study, even if such consents are obtained. Besides, as mentioned earlier, in the view of the author this would still require a very significant expansion of the current paper going beyond the scope of a brief section or sub-section, which could shift the focus away from its original goal as a "Review" paper.
- In the systematisation in the current study, the included studies were categorised into 8 main methodologies, and several of these seperately contain a large number of different methods and approaches on their own. Even if a few concrete examples would be described in sufficient detail, only a fraction of the relevant methods from the presented systematisation could be covered in this way.
As a result of the previous considerations, the inclusion of a concrete example was eventually not implemented. However the author believes that the newly added section describing the advantages of the combination of Discrete-Event Simulation and Deep Learning, together with the developed systematisation provided in Section 4 could adequately and significantly aid the reader in unterstanding the benefits of the tecnology and could also provide a sufficient starting point for dwelling deeper into the intricacies of an actual method or methods in which the reader is interested in.
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors have revised and improved the manuscript based on the given comments. Hence, I have no further comments.
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors have significantly improved the article based on the reviewers' recommendations. This study contributes to the development of a new methodological approach based on the combination of discrete-event simulation and deep learning in logistics. Within this approach, a systematization of literature related to the joint use of discrete-event simulations and deep learning for logistics system optimization was carried out. These goals were achieved through a comprehensive review of relevant literature. The paper is recommended for publication.
