A Three-Layered Multifactorial Evolutionary Algorithm with Parallelization for Large-Scale Engraving Path Planning
Round 1
Reviewer 1 Report
- In page 3 it is stated that ‘The engraving machine studied in this paper is a modification of 3D printer, which was the designed in the literature [20], as shown in Fig.1’. Please correct to ‘The engraving machine studied in this paper is a modification of 3D printer, which was designed in the literature [20], as shown in Fig.1’.
- It is stated in page 4 that ‘…the motor of the X or Y axis should be terminated’. What does it mean terminated? Is it stopped?
- It is stated in page 4 that ‘Where, Eq. (1-1),(1-2) and (1-3) show that the consumed time of the first and second movement routes are equal, which are less than the third movement route’. It is not clear to mw how the first and second route have the same time since the distance is different.
- In section 3.1 the three engraving methods are presented but it is not mentioned which one is selected in this paper and why it is selected.
- It is stated in the introduction that ‘We formulated the trajectory optimization problem in the large-scale image engraving as an improved model based on TSP’. However, this formulation is not presented in the paper.
- The results are a bit confusing. TSP and cities terms are used on images and it is a bit confusing.
- It is stated in the abstract that ‘In particular, the proposed parallel algorithm reduces 80% engraving time. Moreover, the results demonstrated that the engraving time of our algorithm is approximately 1/13 of traditional engraving method’. However, it is not demonstrated anywhere that the proposed parallel algorithm reduces 80% engraving time.
- Please revise the English of the paper. Some sections need to be rewritten in a better way.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The article is aimed at the optimization of laser engraving speed of larger areas without loss of quality, using optimization algorithms based on TSP.
The proposed innovative procedure of a three-layered algorithm based on clustering of the area for engraving enabling parallelization of tasks demonstrates favourable results in comparison with previous algorithms even when using commonly available computational technology.
The introductory parts of the article provide a sufficient overview of the status of the solution to the problem, identification of the problem and an overview of the approach to its solution. The description of the model is presented clearly.
My comments:
1. Abbreviations must be defined when they first occur. E.g. The travelling salesman problem (TSP) is defined in lines 29-30, but the abbreviation is used in line 24 already.
2. It is not clear for which sizes of engraved surfaces the given algorithm is intended - it is necessary to specify a large-scale object.
3. In Table 5, a shorter time is achieved on a conventional computer when calculating the last and largest earring pattern as compared to a supercomputer. How do the authors justify that?
4. Figure 5 shows 2 longer linear "scratches", one of which is at the head of Mony-Lisa, and also a larger black spot next to the neck. Are image errors related to the engraving algorithm?
5. It is necessary to present the resulting image achieved by other compared methods in order to assess their quality, as quality is one of the factors that the authors claim to improve. The quality of the optimized engraving result can be defined in various ways - the approach presented in line 233 is only one option.
Moreover, several typos in the text change the meaning of the words (such as the last word in line 67), and stylistics is also not enough, so grammatical corrections are necessary.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The paper was strongly improved, and questions were explained on a sufficient level. In my opinion, this article brings an inspirative algorithm that would interest readers of the journal.
