Condition-Based Maintenance in Aviation: Challenges and Opportunities
, Bruno F. Santos 2, Floris Freeman 3, Paul van Kessel 3, Dimitrios Zarouchas 4, Theodoros Loutas 5, Richard C. K. Yeun 1 and Iryna Heiets 1Round 1
Reviewer 1 Report
Aerospace-2565120 Condition-Based Maintenance in Aviation: Challenges and Opportunities
This paper comprehensively studies various aspects of the Condition-based Maintenance (CBM) concept in the aviation industry. Specifically, challenges, limitations, and solution directions are discussed from a multi-stakeholder perspective. I recommend that this paper be considered for publication if the authors could address my concerns.
1. The authors might need to include a table or appendix to list and explain the abbreviations in this paper. For example, Line 478, “ACARS” is not explained when it first occurs.
2. 5.1. Data quantity and quality The authors propose leveraging machine learning (ML) models as one future direction. Are there existing relevant studies (or industry applications) that demonstrated the capability and reliability of this method? How about other data analytic methods than ML? (e.g., traditional statistical methods)
3. 5.5. Workforce considerations. The authors indicate that aircraft maintenance technicians and engineers must be trained with the appropriate knowledge regarding CBM. Are there any existing projects/studies from industry manufacturers / academic collegiate aviation training programs that have already started relevant training? If so, please identify it and cite it.
Comments for author File: 
Comments.pdf
Author Response
This paper comprehensively studies various aspects of the Condition-based Maintenance (CBM) concept in the aviation industry. Specifically, challenges, limitations, and solution directions are discussed from a multi-stakeholder perspective. I recommend that this paper be considered for publication if the authors could address my concerns.
Response: thank you for your kind comments on the overall paper. Our response to the individual comments is listed below.
- The authors might need to include a table or appendix to list and explain the abbreviations in this paper. For example, Line 478, “ACARS” is not explained when it first occurs.
Response: Thank you for pointing this out. We have added a list of acronyms / abbreviations in a newly added Appendix A (see tracked change at the end of the paper).
- 5.1. Data quantity and quality: The authors propose leveraging machine learning (ML) models as one future direction. Are there existing relevant studies (or industry applications) that demonstrated the capability and reliability of this method? How about other data analytic methods than ML? (e.g., traditional statistical methods)
Response: Relative to section 5.1, the use of ML models is mentioned relative to two points: the use of ML enabled via Federated Learning, enabling the collection and use of data across individual clients while ensuring data security and ownership. The other use of ML mentioned is in relation to synthetic data development using data augmentation to address the issue of degradation / failure data scarcity. We assume the reviewer’s point is relative to the latter aspect. In this context, there are some recent examples of data augmentation for CBM (comprising SHM and PHM) applications, for instance the references given below, which have been included into the paper (see tracked changes, refs 42-45):
Changchang Che, Huawei Wang, Minglan Xiong, Shici Luo, Few-shot fatigue damage evaluation of aircraft structure using neural augmentation and deep transfer learning, Engineering Failure Analysis, Volume 148, 2023, 107185, ISSN 1350-6307, https://doi.org/10.1016/j.engfailanal.2023.107185.
Dabetwar, S., Ekwaro-Osire, S., and Dias, J. P. (August 23, 2021). "Fatigue Damage Diagnostics of Composites Using Data Fusion and Data Augmentation With Deep Neural Networks." ASME. ASME J Nondestructive Evaluation. May 2022; 5(2): 021004.
Seokgoo Kim, Nam Ho Kim, Joo-Ho Choi, Prediction of remaining useful life by data augmentation technique based on dynamic time warping, Mechanical Systems and Signal Processing, Volume 136, 2020, 106486, ISSN 0888-3270, https://doi.org/10.1016/j.ymssp.2019.106486.
DeQun Zhao and JiaYu Zhao "Remaining life prediction of turbofan engine based on multi-path feature fusion", Proc. SPIE 12257, 4th International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2022), 122573O (1 August 2022); https://doi.org/10.1117/12.2640207
- 5.5. Workforce considerations. The authors indicate that aircraft maintenance technicians and engineers must be trained with the appropriate knowledge regarding CBM. Are there any existing projects/studies from industry manufacturers / academic collegiate aviation training programs that have already started relevant training? If so, please identify it and cite it.
The authors are not aware of any studies or projects from manufacturers and training programs specifically targeting CBM training and as such, we have not included specific references. Having said that, numerous contributing elements are covered in existing programs at technician (LAME) and engineering levels; for instance, several curricula spend time on non-destructive inspection / testing (NDI / NDT) use, or developing, testing and applying machine learning models which can be translated to CBM applications.
Reviewer 2 Report
The research is comprehensive and of a very high standard. It covers an exceptionally significant topic. Several issues should be improved to increase the quality of the paper. After improvements, I highly recommend acceptance of the paper.
1) In lines 228-232 and 233-240, the lack of research in 2 areas is pointed out. The Authors highlight that few papers focused on those problems were published. Please cite those papers.
2) In line 381, the Authors mention limitations in data gathering, transfer and storage capabilities at some locations. Please explain what are the reasons for those limitations. Can the Authors additionally state how often such problems occur? For example, does it happen in the case of 1% or 5% of flights or flying routes? Such a piece of information can highlight the significance of the issue.
3) In point 4.6.1, the economic aspects of implementing CBM are discussed. Please give some approximate value of how the application of the method can influence the current cost structure. The problem is complex, so if the Authors do not have precise data, please consider giving range values depending on assumed limitations.
There are also edition-related issues:
1) There is extra space in line 141 before “In some” and in line 377 after “available”.
2) Point 2.2 is at the bottom of the page and therefore separated from its initial paragraphs. A similar problem is with point 3.
3) I recommend increasing the font size in all figures.
4) The text is not justified in lines 367-432 and 590-598.
Author Response
The research is comprehensive and of a very high standard. It covers an exceptionally significant topic. Several issues should be improved to increase the quality of the paper. After improvements, I highly recommend acceptance of the paper.
Response: thank you for your helpful comments on the overall paper. Our response to the individual comments is listed below.
1) In lines 228-232 and 233-240, the lack of research in 2 areas is pointed out. The Authors highlight that few papers focused on those problems were published. Please cite those papers.
Response: thank you for your helpful comments on the overall paper. Our response to the individual comments is listed below.
2) In line 381, the Authors mention limitations in data gathering, transfer and storage capabilities at some locations. Please explain what are the reasons for those limitations. Can the Authors additionally state how often such problems occur? For example, does it happen in the case of 1% or 5% of flights or flying routes? Such a piece of information can highlight the significance of the issue.
Response: thank you for this comment. We have added the following reasons (see also tracked changes): “outstations in particular may not have sufficient provisions for data transfer (e.g., by not having wireless capacities such as gatelink) or the time and personnel required to facilitate data transfer and storage (e.g., when working with short turn-around times)”. Relative to the frequency of such problems, we have tried to find literature or technical reports quantifying this, but no recent estimates have been found.
3) In point 4.6.1, the economic aspects of implementing CBM are discussed. Please give some approximate value of how the application of the method can influence the current cost structure. The problem is complex, so if the Authors do not have precise data, please consider giving range values depending on assumed limitations.
Response: we have included some quantification of anticipated benefits of CBM in the aviation context, see tracked changes in line 604-606: “Some estimates indicate that CBM may lead to cost savings of up to 700 million euro per year for the European aviation sector alone [ref nr].“
[ref nr] European Commission – Cordis, Horizon 2020 Real-time Condition-based Maintenance for Adaptive Aircraft Maintenance Planning. Accessed online [16-08-2023]: https://cordis.europa.eu/project/id/769288/
There are also edition-related issues:
1) There is extra space in line 141 before “In some” and in line 377 after “available”.
2) Point 2.2 is at the bottom of the page and therefore separated from its initial paragraphs. A similar problem is with point 3.
3) I recommend increasing the font size in all figures.
4) The text is not justified in lines 367-432 and 590-598.
Response: thank you for highlighting these points. We have fixed them in the revised version of the paper, as per tracked changes in the associated lines. Figures have been increased in size, hopefully resolving any readability issues regarding font size. If not sufficient, we are of course more than willing to update the figures to increase font sizes, though this will incorporate some redesign to ensure visual formatting remains acceptable.
