Bacterial Memetic Algorithm for Asymmetric Capacitated Vehicle-Routing Problem

Round 1

Reviewer 1 Report

The paper discusses the Bacterial Memetic Algorithm for Asymmetric Capacitated Vehicle Routing Problem, from the begining it sounds an interesting problem but there is no novelty to the paper. The paper requires substantial revision and enrichment prior to the publication. In its current form cannot be accepted.

1: there is a complete lack of literature review. Which means that there is not scientific background for the proposed method. The travel saleman problem has been extensively been analysed. Authors should carefully read the literature and identify the research gap.

2: currently the paper does not add anything to the literature

3: there should be a comparisson with other methods for solving this problem in order to demonstrate the superiority of the presented one.

4: results are poorly presented

5: there are no strong discussion points

6: reference list is very limited

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper entitled 'Bacterial Memetic Algorithm for Asymmetric Capacitated Vehicle Routing Problem' proposes an adaptation of the Bacterial Memetic Algorithm to the Capacitated Vehicle Routing Problem. I believe the article has some potential, but many things need to be improved in it. My comments:

1. First of all, the article must undergo a linguistic correction (e.g. some abbreviations are missing, once you write 'bacterial memetic algorithm' and sometimes you write 'Bacterial Memetic Algorithm').

2. I believe that references to literature should be removed from the abstract.

3. At the end of the introduction, you should write down the structure of the article (what each section contains).

4. Section 2 lacks a mathematical formulation of the problem.

5. I think that before the description of crossbreeding in section 3 there should be a block diagram of the genetic algorithm (you describe what happens in the crossover step, but it does not appear from the description that it is in the genetic algorithm).

6. The pseudocode of the proposed solution is missing.

7. Nondeterministic algorithms should be executed at least 10 times with a different seed of the pseudorandom number generator (for statistical tests at least 30 times). How many times have you repeated your calculations (e.g. for ABEFMP)? There is no average result, standard deviation, and no statistical tests comparing the results of different variants and algorithms.

8. Values on the X-axis in Figures 5-7 are missing.

9. The name and surname of the author 'AnandSubramanian' in position 3 of the bibliography have been incorrectly spelled.

10. When describing the ABEFMP set, you refer to the paper [A].

11. According to it, ABEFMP instances have been resolved optimally and the paper points to the [B] website, according to which the results of your algorithm, which are presented in table 1, are impossible to achieve (only the results from the literature on the subject with which you compare have the value of the objective function optimal or worse).

 

[A] Uchoa, E.; Pecin, D.; Pessoa, A.; Poggi, M.; Vidal, T.; Subramanian, A. New benchmark instances for the Capacitated Vehicle Routing Problem. European Journal of Operational Research 2017, 256. doi:10.1016/j.ejor.2016.08.012.

[B] http://vrp.atd-lab.inf.puc-rio.br/

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Authors have addressed many of the commants raised by the reviewers, but still the novelty of the paper is not clear. As said in my original review report the problem that is studed in this paper has been significantly addressed in the literature. In order for demonstrating the usefulness and superiority of the propsed method, authors should compare their method with some of the most used methods for this problem, otherwise the reader cannot know which algorithm is the best. If the algorithm is not better from the existing ones there is no need for its existence.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

My comments were duly taken into account, therefore I recommend to accept the paper.

Author Response

Thank you for your positive feedback.