Prediction of Shipping Cost on Freight Brokerage Platform Using Machine Learning

Round 1

Reviewer 1 Report

1. Summary

This study derives the main variables that influence the setting of shipping costs and presents the recommended shipping cost given by a price prediction model using machine learning methods. Thus, this paper could be used to set shipping costs on freight brokerage platforms and to improve utilization rates. However, I am afraid I can only recommend the manuscript for publication if the following concerns are clearly explained.

2. Major concerns

The contribution of the paper is not clear enough; for example, in line 40, "This paper proposes a machine learning-based shipping cost prediction method for a domestic freight transport environment using data from a freight brokerage platform." However, this paper does not propose an innovative model. In addition, the author said that compared with the previous algorithm, environmental factors such as precipitation are included to derive factors that affect how shipping costs are set. However, the experimental results show that environmental factors have no contribution to model prediction, which is contradictory to the contribution claimed by the paper.

3. Minor concerns

A. There are some grammatical errors in the manuscript, and the author is advised to double-check the manuscript.

B. In order to support subsequent follow-up research, it is best for the author to disclose the data sets and codes used in this paper.

C. The parameters of model training should be clearly provided, such as the learning rate, to support the reproducibility of the paper.

D. The picture caption should be placed on the same page as the picture.

Author Response

We appreciate your elaborate comments, which were very helpful in improving our paper. We modified the paper according to your comments and added some additional content. We marked the changes in blue. We hope that the revised manuscript be better with the following modifications.

Author Response File: Author Response.pdf

Reviewer 2 Report

Machine learning-based shipping cost prediction proposed in this paper is considered a very interesting topic. Using the standardized freight rate presented in this study is expected to help sustain business by reducing conflicts between consignors and vehicle owners.

 Please revise the manuscript considering the following review comments.

 - It is necessary to explain the difference between this study and the existing studies on machine learning-based price prediction models more clearly. In particular, the academic contributions of this study should be described more clearly.

- In Section 1. Introduction, the data preprocessing process is explained in detail. It is desirable to cover this in Section 3. Instead, it is necessary to emphasize the background and necessity of this study more clearly in the introduction.

- Figure 1 consists of three images. It is ambiguous which image is after applying preprocessing. Like the caption of the first image, the caption of the second image and the third image are also required to clearly describe that each image shows an image after applying a pre-processing method.

- It is necessary to clearly explain whether the proposed method can be used to determine prices in actual practice. If there is a limit to actual application, it is necessary to describe the related issues more clearly.

- Although the authors mentioned in the conclusion, the analysis was performed by combining the environmental factor data, but no valid results were obtained. So, additional analysis plans for this need to be added in the future research plan. In addition, it is expected that the quality of research results can be further improved by supplementing the method of removing or replacing missing values in the future.

- It is recommended that English presentation should be further improved.

Author Response

We appreciate your elaborate comments, which were very helpful in improving our paper. We modified the paper according to your comments and added some additional content. We marked the changes in blue. We hope that the revised manuscript be better with the following modifications.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The manuscript may be published in its current form.

Author Response

According to the comments of Academic Editor, we have added four references and related information to Section 2 (Literature Review) and Section 6 (Future Research). We would like to thank the reviewers for their efforts to improve the quality of this paper.