Semantic-Based Multi-Objective Optimization for QoS and Energy Efficiency in IoT, Fog, and Cloud ERP Using Dynamic Cooperative NSGA-II
Round 1
Reviewer 1 Report
In the abstract, the background knowledge on the problem addressed needs to be added to some more extent.
In the abstract, the wide range of applications and their possible solutions need to be added in a summarized way.
In the abstract, the problem addressed needs to be justified with more details.
In the abstract, the performance should be described in some numerical value in terms of accuracy, precision, recall, etc.
In the Introduction section, the drawbacks of some conventional techniques should be described clearly.
The introduction section can be extended to add the issues in the context of the current work.
The introduction section should contain the research gaps in summarized form.
There should be different subsections like motivations, contribution, and organization in the Introduction section.
Literature review techniques have to be strengthened by including the issues in the current system and how the author proposes to overcome the same.
There is a lack of current literature in the related work section. It should consist of more related techniques and authors’ findings related to the current work.
Research gaps, objectives of the proposed work should be justified before the problem formulation section.
The authors should consider more recent research done in their study (especially in the years 2018 to 2021 onwards). I have found very weak literature related to the current domain.
It much be better if authors could create an abbreviation table separately.
There should be a different section of problem formulation and a different section related to the proposed solution.
The proposed work should be described with the help of a flowchart and algorithm.
The paper needs to provide significant experimental details to correctly assess its contribution: What is the validation procedure used?
An error and statistical analysis of data should be performed. Further, performance in terms of precision, accuracy, recall and other related parameters should be considered.
Kindly provide several references to substantiate the claim made in the abstract (that is, provide references to other groups who do or have done research in this area).
An error and statistical analysis of data should be performed. Further, performance in terms of precision, accuracy, recall and other related parameters should be considered.
The conclusion should state the scope for future work.
Discuss the future plans concerning the research state of progress and its limitations.
Minor English editing required.
Author Response
We would like to thank the reviewers for their consideration of our manuscript and their positive feedback and their constructive critic. We have made amendments in many parts of the manuscript to fulfil in the best the recommendations and the remarks of the reviewers.
Reviewer #1:
- In the abstract, the background knowledge on the problem addressed needs to be added to some more extent.
We introduced a sentence at the beginning of abstract by providing a brief explanation of the background knowledge of the problem.
- In the abstract, the wide range of applications and their possible solutions need to be added in a summarized way.
Some examples of applications and their possible solutions have been introduced in the abstract.
- In the abstract, the problem addressed needs to be justified with more details.
More details about the problem are added in the abstract. Please refer to sentence in the abstract: “This means that …. “.
- In the abstract, the performance should be described in some numerical value in terms of accuracy, precision, recall, etc.
The requested numerical values are added in the abstract based on two main metrics: quality of service and energy consumption ratio.
- In the Introduction section, the drawbacks of some conventional techniques should be described clearly.
Conventional techniques are not efficient and flexible enough to support service failure and none of them could get an optimum consensus between service quality, service speed and service cost, continuous need changes of customer and limited device’s energy, etc. The paragraph is added in Introduction section after, please refer to that section.
- The introduction section can be extended to add the issues in the context of the current work.
We extended the Introduction with some issues in the context of smart ERP, semantic web technology, multi-agents, and smart sensors, especially dealing with process system fragmented in which multi-agents struggle to communicate and collaborate effectively. However, we can add other issues such as smart real-time production and security issues.
- The introduction section should contain the research gaps in summarized form.
We took this into account by summarizing the research gaps and adding it as new subsection in the Introduction section.
- There should be different subsections like motivations, contribution, and organization in the Introduction section.
We took this into account by adding the research gaps, motivations and contributions, and organization subsections in the Introduction section as requested.
- Literature review techniques have to be strengthened by including the issues in the current system and how the author proposes to overcome the same.
Our work combined several research fields such as ontology, multi-agent, NSGA-II, IoT and Fog. To the best of my knowledge, none of existing works perform Semantic-Based Multi-Objective Optimization QoS and Energy Efficiency in IoT, Fog, and Cloud ERP Using Dynamic Cooperative NSGA-II.
- There is a lack of current literature in the related work section. It should consist of more related techniques and authors’ findings related to the current work.
We have cited and added more recent works from [39, 41, 42] of years 2022 – 2023 in the related work section.
- Research gaps, objectives of the proposed work should be justified before the problem formulation section.
We have added research gaps and objectives at the beginning of the proposed work in the problem formulation section.
- The authors should consider more recent research done in their study (especially in the years 2018 to 2021 onwards). I have found very weak literature related to the current domain.
More recent works are added in the Related Work section. Please refer to lines 71-74, 294, 296, 308-311.
- It much be better if authors could create an abbreviation table separately.
We added abbreviations table in the subsection Abbreviations of the section Proposed Approach.
- There should be a different section of problem formulation and a different section related to the proposed solution.
We separated the problem formulation and proposed solution in different sections: section 5- Problem Formulation and Section 6- The Framework.
- The proposed work should be described with the help of a flowchart and algorithm.
The proposed work is helped by two algorithms in pages 22, 23 and a flowchart of Dynamic semantic-based multi-agent services composition in page 20.
- The paper needs to provide significant experimental details to correctly assess its contribution.
The experimental details are presented in section 6 in terms of three criteria: accuracy solution comparison, performance comparison and energy consumption comparison.
- What is the validation procedure used?
The effectiveness and energy saving of our proposal has been validated and evaluated through multiple experiments on random data sets. To show how our validation works, we illustrate its validation procedure as follows: First, the user specifics his preferences in terms of QoS and required functionalities. Then all performance metrics are evaluated and recorded by different concrete configurations on three multi-agent models: ring, master/slave, and hybrid using simulations of different numbers of services. Finally, these simulations include results obtained from four performance metrics (violation QoS degree, normalized average execution time, average energy consumption rate, and cost) with different multi-agent models through several simulated configurations.
This paragraph is added on lines 731 – 742.
- An error and statistical analysis of data should be performed. Further, performance in terms of precision, accuracy, recall and other related parameters should be considered.
These parameters are used in pattern recognition, information retrieval, object detection, and classification (i.e., machine learning). In our work neither of these domains is handled. Our work focus on optimization field that evaluates the proposed approach in terms of solution quality, QoS and energy violation degree, execution time, number of exchanged messages, and energy consumption comparison in dynamic and adaptive cases.
- The conclusion should state the scope for future work.
We added the scope for future work at the end of the conclusion section.
- Discuss the future plans concerning the research state of progress and its limitations.
We added discussions in the last sentence of the conclusion section.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper represent a semantic dynamic cooperative service selection and composition approach with maximization of customer non-functional needs aimed at quickly selecting the relevant service drive while saving energy. The area of the present research is of great interest in recent years, in particular the study of service compositions and QoS compositions in service systems is included in many international research projects. The abstract is well-written and summarized the results obtained in the paper.
The Introduction explains the basic notion used in the paper such as IoT, Enterprise Resource Planning (ERP). The section is extensive but some of the notions should be elaborated in more details. For example, I recommend to the authors to include a separate paragraph before line 55 which elaborates the notion of QoS and adding some references to it, for example the documents of the International Telecommunication Union for example the QoS regulations (ITU-T Supp. 9 of E.800 Series), QoE requirements for real-time multimedia services over 5G networks (https://www.itu.int/pub/T-TUT-QOS-2022-1); ITU-T Recommendation ITU-T P.10/G.100 (11/2017). Vocabulary for performance, quality of service and quality of experience, etc.
The paragraph between lines 55 and 75 which elaborates the QoS aware service selection and service compositions should include more recent studies on QoS of service compositions.
In particular, an important work on QoS-aware service composition is the paper:
Strunk, A. QoS-aware service composition: A survey. In Proceedings of the 2010 Eighth IEEE European Conference on Web Services, 1–3 December 2010; pp. 67–74.
Another important direction of research recently is the QoS of service compositions under uncertainty which is an important direction of research in recent years. First, in the paper:
Poryazov, S. et al. Two Approaches to the Traffic Quality Intuitionistic Fuzzy Estimation of Service Compositions. Mathematics 2022, 10, 4439. https://doi.org/10.3390/math10234439.
a novel approach to the QoS estimation of service compositions based on the notion of intuitionistic fuzzy sets is proposed. The importance of the proposed approach is due to the fact that it can be applied to all types of service systems.
Another more recent paper on the subject that should be included in the Related work section is again from an MDPI journal – Symmetry:
Tarawneh, H.; Alhadid, I.; Khwaldeh, S.; Afaneh, S. An intelligent cloud service composition optimization using spider monkey and multistage forward search algorithms. Symmetry 2022, 14, 82.
The above sources absolutely should be included in the Introduction in the paragraph between lines 55 and 75 or in the Related works section (Section 3).
The contributions of the paper are clearly stated at the end of the section Introduction. They are excellently formulated.
A motivated case study is described in Section 2. The case study is based on the production and transport business processes of IRIS Company of SETIF city, Algeria. A graphical representation of a Cloud-based ERP production business process for IRIS is shown in Figure 1. The Figure should be resized as the text in the figure is barely readable. The same is true for the graphical representation of the proposed IoT Cloud-based ERP production business process for IRIS.
The authors have correctly stated (paragraph beginning at line 162) that services can be heterogeneous and therefore there will be different data formats. Furthermore, different types of protocols might be used. One additional problem that arises here which the authors have not mentioned is the estimation of QoS of various subservices. The conceptual models should satisfy certain conditions. There is a recent work on overall model normalization towards adequate prediction and presentation of QoS/QoE in overall telecommunication systems where where the importance of the normalization of the concepts, normalization of the indicators’scales, etc. are discussed. These conditions allow for the scalability of the models. This should be mentioned in the last paragraph of Section 2.
The Related works section is well-written and all cited sources are up-to-date and on topic. The alternative approaches are compared in Table 1. The table is of great value and very illustrative.
Section 4 deals with an extension of CxQSCloudSERP Ontology model. Figure 3 should be resized. The intervals on lines 357 and 358 should have a closing bracket at the end – “]” and should occupy the same line as the text before them.
Equation (2) should end with a “.”. Figures 4, 5, 6 should be explained. It is very hard for the reader to understand the models currently.
The Methodology and Framework are very well explained in Section 5. I have no major critical remarks to it, other than the problems with equation formatting. Equations should end with “.” when they appear at the end of the sentence. Before if clause in equation environment there should be “,” (see equations 5,7,8 etc.).
Table 2 has been split into two. The text should be revised so that the last row appears on page 17.
The caption of Figure 10, should be moved to the previous page.
Equations 11 and 12 have the above mentioned formatting issues.
The two proposed algorithms seem correct.
I do not see any problems with the numerical results. They confirm the apriori expectations.
The conclusions drawn by the authors are supported by the results. The last sentence of the conclusion between lines 786 and 788 makes little sense to me. I recommend to the authors to revise this sentence and state more clearly what they mean.
Overall, the paper represents some significant results and is of a very high quality. I recommend that the paper be published once the authors address adequately my remarks.
The language is fine. There are very few mistakes/typos.
Author Response
Responses to reviewers
We would like to thank the reviewers for their consideration of our manuscript and their positive feedback and their constructive critic. We have made amendments in many parts of the manuscript to fulfil in the best the recommendations and the remarks of the reviewers.
Reviewer #2:
- The Introduction explains the basic notion used in the paper such as IoT, Enterprise Resource Planning (ERP). The section is extensive but some of the notions should be elaborated in more details. For example, I recommend to the authors to include a separate paragraph before line 55 which elaborates the notion of QoS and adding some references to it, for example the documents of the International Telecommunication Union for example the QoS regulations (ITU-T Supp. 9 of E.800 Series), QoE requirements for real-time multimedia services over 5G networks (https://www.itu.int/pub/T-TUT-QOS-2022-1); ITU-T Recommendation ITU-T P.10/G.100 (11/2017). Vocabulary for performance, quality of service and quality of experience, etc.
These works are used for QoS and QoE of phone transmission quality, phone installations, Local Line Networks while the proposed approach consists of realizing an optimal and dynamic smart business process based on IoT, Fog, and Cloud ERP which makes the integration in question difficult. This can be justified by the nature and issues of the reference in terms of QoS, QoE of phone transmission quality, phone installations, etc.
- The paragraph between lines 55 and 75 which elaborates the QoS aware service selection and service compositions should include more recent studies on QoS of service compositions.
- In particular, an important work on QoS-aware service composition is the paper: Strunk, A. QoS-aware service composition: A survey. In Proceedings of the 2010 Eighth IEEE European Conference on Web Services, 1–3 December 2010; 67–74.
- We added another important direction of research recently is the QoS of service compositions under uncertainty which is an important direction of research in recent years. First, in the paper: Poryazov, S. et al. Two Approaches to the Traffic Quality Intuitionistic Fuzzy Estimation of Service Compositions. Mathematics 2022, 10, 4439. https://doi.org/10.3390/math10234439.
- A novel approach to the QoS estimation of service compositions based on the notion of intuitionistic fuzzy sets is proposed. The importance of the proposed approach is due to the fact that it can be applied to all types of service systems.
- Another more recent paper on the subject that should be included in the Related work section is again from an MDPI journal – Symmetry:
Tarawneh, H.; Alhadid, I.; Khwaldeh, S.; Afaneh, S. An intelligent cloud service composition optimization using spider monkey and multistage forward search algorithms. Symmetry 2022, 14, 82.
The above sources absolutely should be included in the Introduction in the paragraph between lines 55 and 75 or in the Related works section (Section 3).
This work is added in the Related Work section on lines 308-311 and references.
- A motivated case study is described in Section 2. The case study is based on the production and transport business processes of IRIS Company of SETIF city, Algeria. A graphical representation of a Cloud-based ERP production business process for IRIS is shown in Figure 1. The Figure should be resized as the text in the figure is barely readable. The same is true for the graphical representation of the proposed IoT Cloud-based ERP production business process for IRIS.
Figures 1 and 2 are enhanced and resized.
- The authors have correctly stated (paragraph beginning at line 162) that services can be heterogeneous and therefore there will be different data formats. Furthermore, different types of protocols might be used. One additional problem that arises here which the authors have not mentioned is the estimation of QoS of various subservices. The conceptual models should satisfy certain conditions. There is a recent work on overall model normalization towards adequate prediction and presentation of QoS/QoE in overall telecommunication systems where the importance of the normalization of the concepts, normalization of the indicators’ scales, etc. are discussed. These conditions allow for the scalability of the models. This should be mentioned in the last paragraph of Section 2.
We have mentioned that the estimation of QoS of different services should meet comprehensive model normalization for appropriate prediction and management of QoS in ERP.
This paragraph is added in the last paragraph of Section 2.
- Section 4 deals with an extension of CxQSCloudSERP Ontology model. Figure 3 should be resized. The intervals on lines 357 and 358 should have a closing bracket at the end – “]” and should occupy the same line as the text before them.
Figure 3 is enhanced and resized.
Closing bracket means that the interval is closed but, in our model, low and medium semantic values are open right interval because each value should belong to a single interval: ∩Ii = Ø
- Equation (2) should end with a “.”. Figures 4, 5, 6 should be explained. It is very hard for the reader to understand the models currently.
Figures 4, 5, 6 are models of our ontology where each element in circle represents a model class and the edges represent relationships among these classes.
- The Methodology and Framework are very well explained in Section 5. I have no major critical remarks to it, other than the problems with equation formatting. Equations should end with “.” when they appear at the end of the sentence. Before if clause in the equation environment there should be “,” (see equations 5,7,8 etc.).
It’s revised.
- Table 2 has been split into two. The text should be revised so that the last row appears on page 17.
It’s corrected.
- The caption of Figure 10, should be moved to the previous page.
It’s corrected.
- Equations 11 and 12 have the above-mentioned formatting issues.
It’s formatted.
- The conclusions drawn by the authors are supported by the results. The last sentence of the conclusion between lines 786 and 788 makes little sense to me. I recommend to the authors to revise this sentence and state more clearly what they mean.
We agree with the reviewer, we have changed the last sentence of the conclusion section.
Author Response File: Author Response.pdf
Reviewer 3 Report
The presented work describes recent advancements in industrial wireless technology and IoT, which have brought a shift in smart applications. The paper proposes a semantic dynamic cooperative service selection and composition approach that maximizes customer non-functional needs and quickly selects relevant services while saving energy. The study seems interesting for some groups of users. However, the following concerns need to be addressed.
1. I suggest improving the tile of the paper, the authors may consider either of the following.
"Semantic-Based Multi-Objective Optimization for QoS and Energy Efficiency in IoT, Fog, and Cloud ERP using Dynamic Cooperative NSGA-II"
"Intelligent Resource Management for IoT, Fog, and Cloud ERP using Semantic-Based Dynamic Cooperative NSGA-II Algorithm for QoS and Energy Optimization"
2. "We introduce a new QoS-energy violation degree with leverages QoS energy-efficient cooperative mechanism." Improve the sentence; it does not express the correct sense. What is the meaning of "new QoS-energy violation degree "? Also, mention the percentage of improvement done by the proposed method and the name of the comparative methods in the last sentence of the abstract.
3. IoT, ERP, IIoT, and IRIS expand this abbreviation in the text used for the first time.
4. Figure 1 needs to be clarified, and please improve it.
5. How did the authors achieve the Pareto optimality? I did not see any convergence graph. Please justify.
6. "The proposed research work needs to provide: …." In line 183, change the number to Roman i, ii, iii.
7. The caption of Figure 10 spans through the next page. Adjust it.
8. The manuscript would benefit from improvements in academic writing style and should be thoroughly checked for any typographical or grammatical errors.
9. Consider adding a few latest related articles between 2021 to 2023 to set the background of your work.
The manuscript would benefit from improvements in academic writing style and should be thoroughly checked for any typographical or grammatical errors.
Avoid using too much of compound sentences.
Author Response
Responses to reviewers
We would like to thank the reviewers for their consideration of our manuscript and their positive feedback and their constructive critic. We have made amendments in many parts of the manuscript to fulfil in the best the recommendations and the remarks of the reviewers.
Reviewer #3:
- I suggest improving the tile of the paper, the authors may consider either of the following. "Semantic-Based Multi-Objective Optimization for QoS and Energy Efficiency in IoT, Fog, and Cloud ERP using Dynamic Cooperative NSGA-II" "Intelligent Resource Management for IoT, Fog, and Cloud ERP using Semantic-Based Dynamic Cooperative NSGA-II Algorithm for QoS and Energy Optimization"
We adopted the first suggestion: "Semantic-Based Multi-Objective Optimization for QoS and Energy Efficiency in IoT, Fog, and Cloud ERP using Dynamic Cooperative NSGA-II”
- We introduce a new QoS-energy violation degree with leverages QoS energy-efficient cooperative mechanism." Improve the sentence; it does not express the correct sense. What is the meaning of "new QoS-energy violation degree "? Also, mention the percentage of improvement done by the proposed method and the name of the comparative methods in the last sentence of the abstract.
Numerical findings are added in the abstract based on two main metrics: violation of quality of service and energy consumption ratio. We also added comparative methods in the last sentence of the abstract.
- IoT, ERP, IIoT, and IRIS expand this abbreviation in the text used for the first time.
All requested abbreviations are added to the text.
- Figure 1 needs to be clarified, and please improve it.
Figure 1 is clarified and improved in the main manuscript.
- How did the authors achieve the Pareto optimality? I did not see any convergence graph. Please justify.
The optimization algorithm performs the convergence to the non-domination solutions.
Figure 14 shows the Pareto Optimal for Qos score and energy consumption. The red stars in figure 14 represent the near-optimal non domination solutions (Pareto optimality) according to different strategies.
- The proposed research work needs to provide: …." In line 183, change the number to Roman i, ii, iii.
It’s Corrected.
- The caption of Figure 10 spans through the next page. Adjust it.
It’s Adjusted.
- The manuscript would benefit from improvements in academic writing style and should be thoroughly checked for any typographical or grammatical errors.
The paper is proofread.
- Consider adding a few latest related articles between 2021 to 2023 to set the background of your work.
More recent works are added in the Related Work section. Please refer to lines 71-73, 294, 308-311.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
I am satisfied with authors response and accept this manuscript.
Reviewer 2 Report
I would like to thank the authors for taking into account my remarks. The paper has been significantly improved. I recommend that the paper be published.
English language and style are fine. Minor editting is required.
