A Knowledge-Based Cooperative Differential Evolution Algorithm for Energy-Efficient Distributed Hybrid Flow-Shop Rescheduling Problem

Round 1

Reviewer 1 Report

The research presented in the article is interesting and both the subject - Energy-Efficient Distributed Hybrid Flow-Shop Rescheduling Problem and the proposed algorithm - Knowledge-Based Cooperative Differential Evolution with Hybrid Initialization and Local Intensification should be rated highly.

I have a few small comments that could make the article even more complete:

1) Is the EDHFRP problem a problem known from the literature, or do the authors define it as a variant of the DHFSP or other problem? In the literature you can find, for example, Wang, Jing-Jing & Wang, Ling. (2018). A Knowledge-Based Cooperative Algorithm for Energy-Efficient Scheduling of Distributed Flow-Shop. IEEE Transactions on Systems, Man, and Cybernetics: Systems, doi: 10.1109/TSMC.2017.2788879)

If it is new (and you can learn about it in the research part - point 4.6), then when defining the model, it would be worth emphasizing that the authors define a new problem as an extension of an existing one - this will be an additional advantage of the paper. It would be also good to add a citation where similar models are defined (we don't even have a citation for DHFSP).

2) It would be worth mentioning on what basis the authors chose mutation strategies for DE? Was it the result of previous research or e.g. the most popular set of strategies typically used for the FSP problems? The same applies to the crossover operator.

3) When it comes to the execution time of the algorithm, the product of the instance parameters (0.1 × F × n1 × s) limits the time, but it is expressed in seconds? It may also be worth giving an approximate time interval so that the reader does not have to calculate it.

4) The literature review is well done in my opinion, but I would add the one item mentioned above.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper addresses an important and practical problem: the distributed hybrid flow shop rescheduling with the bi-objective of minimizing the makespan and the energy consumption. The dynamic (stochastic) case was considered in which new jobs can arrive during the time horizon. So the orders are split into sets: the known jobs and the newly arriving jobs. It confers a realistic feature to the problem under study.

A new Differential Evolution metaheuristic was proposed to solve the problem. All the conception’s steps of the novel algorithm were described properly in details, including ideas of the literature e new experiments.

The authors described the tuning process of the parameters of the proposed metaheuristic. And results were analyzed by mean of adequate metrics for multi-objective methods.

Experiments comparing diverse versions of the algorithm were conducted (including the variant with random initial population and a different local intensification). Finally, the best version of the metaheuristic was compared to four other algorithms of the literature and it proved to best the most efficient.

A point that can be improved is about the mathematical model presented as no computational or numerical results were demonstrated. Although it is very clear that the objective of the work was to propose a heuristic approach to solve the problem, once the mathematical model is presented, some kind of validation need to be included, such as the computational execution of a small instance and comments about the CPU time required and the dimensions of the model in terms of number of variables and constraints.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf