A Machine-Learning-Assisted Simulation Approach for Incorporating Predictive Maintenance in Dynamic Flow-Shop Scheduling
Round 1
Reviewer 1 Report
Dear authors,
this article is intersting and well written in general; however, please adress my comments given below:
- please provide new version of Figure 8 (very bad quality)
- abstract provides general overview of article; obtained results and data are missing
- please drawn a comparison between your research and the others (at international level)
- provide better description of state of the art, add more significant research in the field of scheduling problem, e .g. Ojstersek, R.; Lalic, D.; Buchmeister, B. A new method for mathematical and simulation modelling interactivity: A case study in flexible job shop scheduling. Advances in Production Engineering & Management, 2019, 14(4), 435-448.; Balog, M.; Dupláková, D.; Szilágyi, E.; Mindaš, M.; Knapcikova, L. Optimization of time structures in manufacturing management by using scheduling software Lekin. TEM Journal, 2016, 5(3), 319-323.; Abdel-Basset, M.; Mohamed, R.; Abouhawwash, M.; Chakrabortty, R. K.; Ryan, M. J. A Simple and Effective Approach for Tackling the Permutation Flow Shop Scheduling Problem. Mathematics, 2021, 9(3);Duplakova, D.; Teliskova, M.; Duplak, J.,;Torok, J.; Hatala, M.; Steranka, J.,; Radchenko, S. Determination of optimal production process using scheduling and simulation software. International Journal of Simulation Modelling, 2018, 17(4), 609-622., etc.
- figure 11 not povides necessary information, in my opinion this figure should be deleted
- please provide a detailed description of the science and practice contribution of this research article
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The whole paper structure is basically good. A thing that can be improved is the title of the paper. By reading it, in my opinion, the reader can not understand that the machine learning methods are used for the PdM - to extract the maintenance time slots - and not in the scheduling process. So consider making the title more precise, in order to make clear that the ML is not used for the scheduling itself, but for the extraction of the PdM time slots, that the scheduling takes into account.
The Figures in the whole paper can be improved in terms of quality. More specific comments are provided for each section.
English language can be further improved, regarding grammar and spelling. More specific comments are provided for each section.
Abbreviations not always mentioned properly. More specific comments are provided for each section.
Introduction
In general it is a well written abstract, although it can be improved based on the general comments. Some additional comments are given below:
- Line 17: “Info & Communication Technology (ICT) is…”
- Lines 23-29: Mentioning more disruption causes than you originally take into account in a way that can mislead the reader. Make clear why are the two ones mentioned in line 29 the most crucial of the others.
- Line 32: “… is implemented by…”
- Line 39: Replace “…factory X scheduling…” with “… factory X production scheduling…”
- Line 40: “…discussion…”
Literature review
- Line 47: to/what where the results compared? Consider writing a line to give more details.
- Lines 59-62: In the previous lines you mentioned the papers’ main idea and results, but in this reference you give a definition of reinforcement learning. The paragraph’s structure till here is the following: paper mention – main idea – results. Try to do the same here. In my opinion, the reinforcement learning definition is not essential for the reader. If you think this is something important though, and you want the reader to learn what reinforcement learning is in a few lines, try to give the definition and the benefits by applying this method. This would be something more useful, in my opinion, for the reader.
- Lines 63-64: TDGA – Two-phase Decoding Genetic Algorithm
- Line 75: flow-shop mentioned here in a different way than before. Try to keep the same form everywhere.
- Lines 82,86: methods with “m” lowercase and genetic algorithm is mentioned here with uppercase initial, while previously was mentioned with lowercase. Try to keep the same form everywhere.
- Lines 95-98: job-shop not mentioned everywhere in this way. Try to keep the same form everywhere.
- Line 99: Try the following: “Although previous works did [this], they did not do [that].”
- Line 101: down-times? Not so clear what this means.
- Lines 101-103, 106: Make sure this sentence is correct regarding English language.
- Line 107: “adding…process”” should be placed right after the word “strategy” and then write the sentence “since…”.
- Line 108: “Predetermined Time-based…”, consider adding an abbreviation for it.
- Line 120: “Artificial Intelligence (AI) tools…”. Also in line 57 AI mentioned as “artificial intelligence”.
- Line 119: “…is the lowest…” or “…is lower…”.
- Line 121: “time slots” instead of “slots”.
Lines 43-128: Abbreviations should only be defined once and in case they are used again later in the document.
- In ‘’Literature Review’’ section you should refer others’ work and methodologies in order to identify important gaps that the present approach deals with. It should be direct to the reader, at the end, what are the gaps and the key contributions of the present work in comparison to the identified gaps of other methods. Maybe you should add more papers regarding production scheduling and maintenance and investigate more the steps that these methods propose:
- https://doi.org/10.1016/j.procir.2018.09.065
- https://doi.org/10.3390/s21030972
- https://doi.org/10.1016/j.jmsy.2021.09.018
- https://doi.org/10.1007/s00170-011-3652-4
- https://doi.org/10.1016/j.ifacol.2016.07.782
Methodology
- Figure 1: better image quality and horizontal labels instead of vertical. Need to improve this figure that explains the framework.
- Figure 2: better quality
- Line 160: here you give a definition of “make span”, although you have mentioned a couple times before.
- Line 166: this sentence’s meaning could be expressed better.
- Lines 167-175: a more formal writing would be better here
- Line 176: How it is decided which data is required to be analysed?
- Lines 199-200: phase seven not so clear/well described
- Line 217: “explained”
- Figure 3: better quality and try to explain better the model. By looking the image, would be better to give more details about the flowchart algorithm, as mentioned in the lines that follow this figure.
- Line 249: “the” or “a”
- Line 270: “need”
- Line 275: Why machine’s vibration, temperature and pressure?
- Figure 4: Enlarge the first part. It is difficult to read letters. Improve graphics.
Case study
- Split Figure 5 graphs between pages 9 and 10. The gap is too big in page 9.
- Figures 5,6,7: Better quality. Highlight with the red rectangle is not so good. Try to make letters BOLD or another color.
- Lines 354-355: the meaning is not so clear.
- Figure 8: better quality.
- Figure 10: blocked, breakdown and setup have similar colors. Also, when blocked happens, as you mentioned, the maintenance is taking place. What happens when a breakdown is predicted
- Figures 9,10: color code is not so easy to read.
Discussion
- Line 390: “DES…”
- Lines 415-416: Improve this sentence.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Dear authors,
thank you very much for your revision. In my opinion, the revised article is suitable for publication in present form.
