Parcel Classification and Positioning of Intelligent Parcel Storage System Based on YOLOv5
Round 1
Reviewer 1 Report
This study looks into how Artificial Intelligence technology can be used in the last part of the logistics industry to deliver parcels without people. Scientists tested different AI technologies and found that one called YOLOv5l worked the best. This study will help set the foundation for using AI technology in delivering parcels.
But before it can be accepted, the author must address several things so that this article can be more precise.
# In the introduction, the author has not explained the advantages and disadvantages of several YOLO versions, the methods and results of which are discussed in detail. Consider presenting the matter in one more paragraph.
# In the introduction, it is crucial to explain to the reader the difference between AI you mean and the deep learning algorithm used in this study to avoid misunderstanding.
# In the introduction, in lines 68-74, the paragraph should be considered for deletion because it is useless to explain this matter.
# In methodology, Figure 2 is a general explanation, so it should be considered for deletion.
# In methodology, Figure 3 is important to improve the resolution to be more readable for readers.
# In experiments, this Figure 5 does not provide good information; consider explaining it in paragraphs.
# In experiments, Figure 6, can the author see this picture? If not, consider replacing it with a better resolution.
# In conclusion, important results should be stated in a short, concise, and clear manner, not beating around the bush. Significant results such as accuracy are the main things the authors do not convey in this conclusion.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The manuscript is entitled "E-Commerce Parcel Detection and Storage System Based on Artificial Intelligence Technology."
The title of the article is very generic. The author needs to mention the novelty of the work by comparing it with existing pretrained models.
What kind of artificial intelligence technology is used to detect the object (parcel)?
There are many considerations to take into account when detecting the object. i.e., edges, dimensions, sizes, etc.), there are many constraints to detect the object. The author needs to mention how the proposed methodology works to detect the parcel based on the storage system.
Technically, it is not obvious; you must prove it mathematically.
The author mentions in the abstract that "among the four detailed models of YOLOv5, YOLOv5l was analyzed to have the best parcel recognition performance, with precision, recall, and F1 scores of 0.966, 0.899, and 17, 0,932, respectively." How do Yolov5 and Yolov5l represent the four models.?
The results are shown in Figure 6. The outcome of the YOLOv5-S, M, L, X... experiment is unclear. need to describe it with technical parameters.
References are not adequate. The author needs to do more literature reviews on pretrained models to detect objects, especially working in warehouse object detection.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The abstract clear and consist of the research objective, brief meth, and summary of the research results. The keywords are also relevant to the research subject. However, the abbreviation appeared in the text for the first needs explanation, (e.g. combines AI technology,....... /and such as GPS and RFID (line 87)), so please double check the whole paper with this regard.
All the elements of the introduction are clearly discussed. However, it could be better by developing the importance for the subject of interest. In addition, the style of citation need to be modified in some statements (e.g. Representatively, CTT in Portugal (CTT, 2020), Correos in Spain (Correos, 2017), La Poste in France (LaPoste, 2016), and Australia (AustraliaPost, 2021) and Hong Kong Post (HongKongPost, 2017), etc. [2] In Korea, deliv......)
The literature review section has been well-structured and detailed with relevant literature included and well presented.
The research methodology is well described.
The implications were not offered. Authors need to identify clearly in the manuscript any implications for research (contributing to the body of knowledge), practice and/or society. The authors could have at least inform the readers the key lesson learnt.
The limitations also not provided, and some work is needed to identify the research limitation correctly. The same applied on the future research, and more detailed description is required.
The study used only 24 research paper. I believe that stander research paper should have at least 30 to 50 references. So please give a credit to other researchers by citing them.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 4 Report
1. In the introductory part of the study, there was a sudden transition from the parcel transportation systems in the world to the logistics industry in Korea. In the study, there is an innovative system proposal for shipping. But is this system only for Korea? Why is the emphasis on Korea? I think this part should be removed. If there is a problem situation, I think it is valid for the transportation sector in all countries of the world. Therefore, it should be defined as a worldwide problem situation.
2. I think that "YOLO v7", which is a faster and up-to-date algorithm version, should be used in the study. In addition, sharing the raw data will contribute to future researchers. In the study, I did not encounter a link that raw data was shared.
3. Some of its researchers also seem to have previous research on cargo transportation, which has also been published in this journal. If researchers are doing this study as a continuation of another existing study, I think it would be better if they could refer to their past research.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
I accept this manuscript. Overall manuscript modified as per review comments.
