A Hybrid Search Using Genetic Algorithms and Random-Restart Hill-Climbing for Flexible Job Shop Scheduling Instances with High Flexibility

Round 1

Reviewer 1 Report

Comments on “A hybrid search using genetic algorithms and random-restart hill-climbing for flexible job shop scheduling problems with high flexibility”

This manuscript presented a new hybrid algorithm called GA-RRHC based on genetic algorithms and a random-restart hill-climbing algorithm for the optimization of flexible job shop scheduling problems with high flexibility. This paper is well-organized and provides some interesting results in some sense. However, the following comments should still be taken into account.

 

(1) In the title of this article, ‘flexible job shop scheduling problems’ is used and shortened to ‘FJSSPs’. However, in the rest of the paper, ‘FJSSP’ is used heavily. The author should make sure that the expression is consistent.

 

(2) There is something wrong with the logic of this article. Content in 3.3, 3.4, 3.5 and 3.6 is a detailed explanation of 3.2. Thus, these sections should be subsection of 3.2.

 

(3) In Experimental results, the author only explained the results in words, which led to the presentation of experimental results not particularly clear and convincing. The results of the experiment can be better illustrated by the appropriate picture presentation.

 

(4) In Comparison with other methods, the author compares the proposed method with other methods to show that the algorithm in this paper has better effect. However, a single textual description seems unconvincing. Appropriate comparison diagrams should be used. What’s more, the author should choose the methods in the same environment for comparison to increase persuasibility.

 

(5) In Reference, some reference titles capitalized the first letter of each word, and some reference titles capitalized only the first letter. Authors should carefully check the format of references to ensure consistency.

 

In conclusion, this paper had given some interesting results. However, the author should provide more comparison studies to illustrate the conclusions, which can thus reflect the advantage of the proposed method. I would like to suggest that it should be carefully revised to address the presented comments.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Problem constraints can be written in mathematical form

Figure 2 is difficult to understand. A better example can be drawn using unequal machines and operations (i.e., 4 operations and 3 machines). 

In mutation operators (3.5), the second mutation is the random change of positions for operations of three different jobs.  Why three jobs?

You employ local search for each smart cell. You may follow the following method (Hybrid Genetic and Simulated Annealing Algorithm for Capacitated Vehicle Routing Problem) to improve the time complexity. 

Experiments for n=6 to 30 and m=5 to 15 are conducted. You may conduct experiments for randomly generated large problem instances. 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf