Hybrid NHPSO-JTVAC-SVM Model to Predict Production Lead Time

Round 1

Reviewer 1 Report

This paper investigates the production lead time prediction problem. A hybrid NHPSO-JTVAC-SVM model is proposed to solve it. Experiments on two datasets collected from a shipyard’s block assembly and a pre-outfitting process have been carried out. Comparisons with other heuristic algorithms are conducted to validate the proposed method.

Overall, it is well-organized research. The writing is good. Some detailed comments to further improve it are as follows.

1. P2，Figure 1 (b), “Delivery time” is not properly shown.
2. P4, it is recommended to separate the following part to be two sentences. “…the radial basis function (RBF, Equation (9)) kernel is better than other kernel functions; therefore, the RBF is used as the kernel function in this study [27].” -> “…the radial basis function (RBF, Equation (9)) kernel is better than other kernel functions. Therefore, the RBF is used as the kernel function in this study [27].”
3. The proposed algorithm is based on PSO. There are many modified PSO variants. It is helpful to extend the review for existing PSO versions. Some useful publications are recommended as examples. (1) Comprehensive learning particle swarm optimization algorithm with local search for multimodal functions. IEEE Transactions on Evolutionary Computation 23 (4), 718-731, 2019. (2) Ensemble particle swarm optimizer. Applied Soft Computing, 55, 533-548, 2017, etc.
4. Figure 6, The arrow linking to “FS” is better to link an accurate process. “FS” can be an explanation for the big box.
5. P9, in the explanation text for Figure 6, the 4^th item should be improved. There are not step numbers in the figure. However, which is “Step 3” as you mentioned in the text?
6. The satisfy condition can be clearer. How to define the desired number of features?
7. How many independent runs have been conducted for each tested algorithm? Please provide detailed information on all key parameters.
8. There are many heuristic algorithms. Why do you choose PSO, GA, BA, and GOA to compare with? Proper citations should be added to them. Key parameters settings for them should be clearly provided.

Author Response

Thank you for your valuable comments.

We tried our best to reflect your comments.

Please check the attached file.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper has a research on Hybrid NHPSO-JTVAC-SVM model to predict production lead time. This paper proposes the new self-organizing hierarchical particle swarm algorithm (PSO) with jumping time-varying acceleration coefficients (NHPSO-JTVAC) - support vector machine (SVM) regression model to increase the accuracy of lead-time prediction by combining the advanced PSO and SVM. The information is interesting and innovative and provides useful insight.

Introduction, subject matters and conclusions are brief, correct and impartial. The organization and structure of the paper are strong.

A detailed comparative analysis of the prediction results of each SVM-based model with those of other conventional machine-learning algorithms is presented.

The references are properly organized to show the gap in the literature. But the references are quite old to gaurantee the originality of the authors` work. It is suggested to cite recent scientific papers which sufficiently covers the area of specific topic.

The paper can be considered for this journal.

Author Response

Thank you for your valuable review.

We have added some more references along with your comments.

Please check the attached file.

 

Author Response File: Author Response.pdf