Next Article in Journal
Investigation of Load Environment and Bending Load Capacities of Aged Prestressed Concrete Sleepers
Next Article in Special Issue
Optimisation of Buffer Allocations in Manufacturing Systems: A Study on Intra and Outbound Logistics Systems Using Finite Queueing Networks
Previous Article in Journal
Pore Structure and Fluid Evaluation of Deep Organic-Rich Marine Shale: A Case Study from Wufeng–Longmaxi Formation of Southern Sichuan Basin
Previous Article in Special Issue
Load Balancing of Two-Sided Assembly Line Based on Deep Reinforcement Learning
 
 
Article
Peer-Review Record

Production Planning Forecasting System Based on M5P Algorithms and Master Data in Manufacturing Processes

Appl. Sci. 2023, 13(13), 7829; https://doi.org/10.3390/app13137829
by Hasup Song 1,2, Injong Gi 1,2, Jihyuk Ryu 1,2, Yonghwan Kwon 3 and Jongpil Jeong 1,*
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Appl. Sci. 2023, 13(13), 7829; https://doi.org/10.3390/app13137829
Submission received: 13 June 2023 / Revised: 29 June 2023 / Accepted: 30 June 2023 / Published: 3 July 2023
(This article belongs to the Special Issue Design and Optimization of Manufacturing Systems)

Round 1

Reviewer 1 Report

Please read the attachment. Thank you.

Comments for author File: Comments.pdf

minor changes are needed.

Author Response

Thank you very much for reviewing the manuscript 
For a point-by-point response to the reviewer's comments, please see the highlights in the attached file "applsci-2476683_reviewer1".
We have followed your suggestion and have revised my manuscript. 
Revisions to the manuscript are shown in blue font and tagged with 'Rev'.
We divided each point for the paper review, and wrote corrections and comments. 
Please read the attachment. Thank you.

Author Response File: Author Response.zip

Reviewer 2 Report

The paper provides a valuable contribution by addressing the challenge of data reliability in smart factories and proposing using the M5P algorithm for predicting dynamic data and improving production planning. However, there are some comments to be worked on.

How to deal with the data diversity of the present moment and moment in the future?

How to ensure the robustness of the model in a highly noisy environment?

Was the tree trained using standard hyperparameters, or were they altered?

Hyperparameters of the tree must be included.

Discuss the effect of n-estimators on the performance of the tree.

Which criterion was used while training the tree? Was it Gini or Entropy? Why?

Did you check the influence of ‘max_depth’ and ‘min_samples_split’ on the testing of the tree? If not, this study is a must. All above comments can be answered by referring to “Augmentation of Decision Tree Model Through Hyper-Parameters Tuning for Monitoring of Cutting Tool Faults Based on Vibration Signatures.”

Comment on computational time and complexity in the training of trees.

Could you provide more information about the training and testing process of the M5P algorithm? What features are used as inputs, and what is the target variable for prediction? How is the accuracy of the model evaluated? It lacks fundamentals of decision tree algorithms to be included referring to “A machine learning approach for vibration-based multipoint tool insert health prediction on vertical machining centre (VMC),” “Application of machine learning for tool condition monitoring in turning.”

The paper mentions that the proposed approach provides indicators to compare production quantity and work time. However, it needs a thorough discussion of these indicators and their significance in production planning.

Include quantitative results and analysis to support the claims made in the paper. How much improvement does the M5P algorithm offer regarding data prediction and production planning?

Check that the abstract provides an accurate synopsis of the paper. It is very vague in its present form.

The methodology of the proposed model must be illustrated by a clear flowchart.

Numerous statements in this paper are untrue and misleading.

Besides, the writing of the paper, including contributions and methodologies, should be clearer and highlight the innovation of methods & principles.

Was the data normalized/ standardized?

The figures could be of better quality, not appropriately labeled, or not cited in the text.

Figure 10. Decision Tree generated by the M5P algorithm using Weka and Figure 11. Linear models created with the M5P algorithm using Weka are redundant, or you would need to improve the presentation by refereeing to the paper mentioned above or removing them.

The overall presentation of results in the form of images and tables is taken as it is from the Weka background, which makes the presentation of the paper poor. You need to improve them.

I’ve suggested a few of the articles just for your reference and hope that these articles direct and guide you in your future work. If you find them worthy and exciting, you may refer them. All the best. Looking to receive a revision of your manuscript.

 

Moderate editing of English language required

Author Response

Thank you very much for reviewing the manuscript 
For a point-by-point response to the reviewer's comments, please see the highlights in the attached file "applsci-2476683_reviewer2".
We have followed your suggestion and have revised my manuscript. 
Revisions to the manuscript are shown in blue font and tagged with 'Rev'.
We divided each point for the paper review, and wrote corrections and comments. 
Please read the attachment. Thank you.

Author Response File: Author Response.zip

Round 2

Reviewer 2 Report

The authors have addressed all my comments positively. I recommend acceptance of the paper.

Except figure 10: M5 unpruned model tree generated by the M5P algorithm using WekaREV. This tree can be redrawn in MS PowerPoint, MS Paint, or any relevant software. Refer to the tree which is drawn in the paper: http://papers.phmsociety.org/index.php/ijphm/article/view/2929

Back to TopTop