Development of Multilayer-Based Map Matching to Enhance Performance in Large Truck Fleet Dispatching

Round 1

Reviewer 1 Report

The authors make a strong effort to improve their article, nonetheless, there are still issues that should be addressed:

a) authors improved the methodology section of the article, but the scientific justification of the suitability of the methods used is still lacking: why were these methods used when more than one option could have been used? (step 3) As an example, lines 573-575 provide information regarding the suitability of one of the options used, but is it provided only in the discussion section and the works that are likely to be referred in the "have been shown to have ..." are not mentioned.

b) please revise the number of the tables mentioned in lines 368 and 369 as they do not seem to match the description

c) table 2 mentions that data was collected from 1000000 mobile car kits. It would be interesting to understand how the car kits were placed and how data reached the researchers and was transformed into a database would be provided

Author Response

Response Reviewer 1:

Comments and Suggestions for Authors

The authors make a strong effort to improve their article, nonetheless, there are still issues that should be addressed:

1. authors improved the methodology section of the article, but the scientific justification of the suitability of the methods used is still lacking: why were these methods used when more than one option could have been used? (step 3) As an example, lines 573-575 provide information regarding the suitability of one of the options used, but is it provided only in the discussion section and the works that are likely to be referred in the "have been shown to have ..." are not mentioned.

Response: Dear Reviewer:

(a.1) We rewrite 4 steps and redraw Figure 1 for the methodology as bellow (Line 224-245).

Since the return of coordinate data from vehicles is continuous, the fleet dispatching management platform may quickly become overloaded. This paper uses a point-to-polygon method to deal with the high volumes and rates of location GPS data returned by large truck fleets. We also used the same technology to process large fleet location data and found that multiple layer grouping combinations can be used to reduce loading. Figure 1 shows four steps of the process to achieve efficiency improvements with a new multilayer-based algorithm for map matching.

Step 1: Initially, there are two procedures for traffic facility and road data. The traffic facility can be constructed a polygon. For road data, center lines using vector data are collected from ArcGIS [37]. These center lines were used to produce buffers, and then the road can be constructed a polygon. For proceeding multilayer-based map matching, we constructed a large fleet management system based on a structured query language (SQL) database for several segments from road network data to process map matching.

Step 2: Truck fleet data were collected from onboard GPS and plane coordinates of TWD 97 TM 2° obtained for sample testing.

Step 3: The puncture (traditional) method (Line249-250) using vector data has more relevant but complex structures were used to partition and compose the digital map. Thereafter, we used an improvement method, adding spatial division index, which still requires complex mathematical procedures. Since more large-scale data are used for reception and cause computational bottleneck, we proposed a look-up table method using raster data to improve efficiency.

Step 4: A comparison of the methods based on the two data types was made to provide guidance for large truck fleet management.

(a.2) (step 3) As an example, lines 573-575 provide information regarding the suitability of one of the options used, but is it provided only in the discussion section and the works that are likely to be referred in the "have been shown to have ..." are not mentioned

Dear Reviewer: We rewrite the sentences as bellow.

This paper focuses on map positioning and efficiency performance. Vector-based and raster-based formats have been shown to have advantages depending on different scenarios and so in this study we used the point-to-polygon approach and then integrated these technologies and methods for improving performance (Line 577-580).

1. please revise the number of the tables mentioned in lines 368 and 369 as they do not seem to match the description (Kevin)

Response b: Dear Reviewer: we rewrite the sentences as bellow.

We consider that traffic facilities consist of interchange and service stations. Roads include national highways, express highways, provincial highways, county highways, district roads, and normal roads. Our large fleet management system includes segment data arranged as in Table 1. Traffic facilities and roads as eight-segment layers on Taiwan main island collected from GPS instruments installed on large trucks. Table 2 shows the quantity (QTY) of GPS data has one million empirical data points. Transmission is the fourth generation of mobile phone mobile communication technology standards (4G). The source data are collected from 35 trucks with eight working hours per day for one month, and the sampling interval is 30 seconds (Line 372-376).

1. table 2 mentions that data was collected from 1000000 mobile car kits. It would be interesting to understand how the car kits were placed and how data reached the researchers and was transformed into a database would be provided.

Response c: Dear Reviewer: We revise the equation (3) and sentences as bellow.

,,

where GPS is the quantity (QTY) of GPS data. TF denotes the quantity of GPS data for each traffic facility and there are i categories of TF. R denotes the quantity of GPS data for each road and there are j categories of R.

The first author is managing the company for commercial trucks, and he collected GPS data from car kits.

Table 2 shows the quantity (QTY) of GPS data has one million empirical data points. Transmission is the fourth generation of mobile phone mobile communication technology standards (4G). The source data are collected from 35 trucks with eight working hours per day for one month, and the sampling interval is 30 seconds (Line 372-376).

Dear Reviewer:

Thank you for your suggestion on how to make our research more efficient. Wish you all the best in new year 2021!

Best regards,

Van

Submission Date

27 November 2020

Date of this review

08 Dec 2020 17:17:28

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper investigates map matching for large scale truck fleet dispatching, which imposes particular challenges for the general map matching problems. Moreover, the authors propose a solution to improve the runtime efficiency of the map matching process.

However, the reviewer finds that the paper suffers from the following issues:

• The English writing needs substantial improvement, as the reviewer finds it challenging to get the critical points in the texts;
• The paper fails to discuss the novelty of the proposed tasks transparently, especially for map matching (a well addressed practical problem);
• For some notations, the corresponding texts fail to deliver a clear explanation (e.g., Eq. 3); 
• Experiments are needed for comparisons between the proposed method and other well-cited map matching methods.

Author Response

Response Reviewer 2

Comments and Suggestions for Authors

The paper investigates map matching for large scale truck fleet dispatching, which imposes particular challenges for the general map matching problems. Moreover, the authors propose a solution to improve the runtime efficiency of the map matching process.

However, the reviewer finds that the paper suffers from the following issues:

• Point 1: The English writing needs substantial improvement, as the reviewer finds it challenging to get the critical points in the texts;
• Response 1: Dear Reviewer: We have used MDPI english editing service, then we asked Professor ME Meadows (native english speaker, President of IGU) for editorial assistance in compiling the final version of the manuscript.
• Point 2: The paper fails to discuss the novelty of the proposed tasks transparently, especially for map matching (a well addressed practical problem);
• Response 2: Dear Reviewer: We rewrite 4 steps and redraw Figure 1 for the methodology as bellow (Line 224-249).

Since the return of coordinate data from vehicles is continuous, the fleet dispatching management platform may quickly become overloaded. This paper uses a point-to-polygon method to deal with the high volumes and rates of location GPS data returned by large truck fleets. We also used the same technology to process large fleet location data and found that multiple layer grouping combinations can be used to reduce loading. Figure 1 shows four steps of the process to achieve efficiency improvements with a new multilayer-based algorithm for map matching.

Step 1: Initially, there are two procedures for traffic facility and road data. The traffic facility can be constructed a polygon. For road data, center lines using vector data are collected from ArcGIS [37]. These center lines were used to produce buffers, and then the road can be constructed a polygon. For proceeding multilayer-based map matching, we constructed a large fleet management system based on a structured query language (SQL) database for several segments from road network data to process map matching.

Step 2: Truck fleet data were collected from onboard GPS and plane coordinates of TWD 97 TM 2° obtained for sample testing.

Step 3: The puncture (traditional) method using vector data has more relevant but complex structures were used to partition and compose the digital map. Thereafter, we used an improvement method, adding spatial division index, which still requires complex mathematical procedures. Since more large-scale data are used for reception and cause computational bottleneck, we proposed a look-up table method using raster data to improve efficiency.

Step 4: A comparison of the methods based on the two data types was made to provide guidance for large truck fleet management.

• Point 3: For some notations, the corresponding texts fail to deliver a clear explanation (e.g., Eq. 3); 

Response 3: Dear Reviewer: We revise the Equation (3) and sentences as bellow. In this paper, we suggest that the constraint of road networks can be divided into two categories, viz. traffic facilities and roads (Line 259-260). According to Equation (3), quantity of traffic facilities plus quantity of roads equals quantity of GPS data from road networks. In section 4, quantity of GPS data is 1000000.

,,

where GPS is the quantity (QTY) of GPS data. TF denotes the quantity of GPS data for each traffic facility and there are i categories of TF. R denotes the quantity of GPS data for each road and there are j categories of R.

• Point 4: Experiments are needed for comparisons between the proposed method and other well-cited map matching methods.

Response 4: Dear Reviewer: The traditional well-cited map matching methods are using vector data (puncture method). We proposed a look-up table for comparison. We rewrite the sentences in Step 3 (Line 238-245).

Step 3: The puncture (traditional) method using vector data has more relevant but complex structures were used to partition and compose the digital map. Thereafter, we used an improvement method, adding spatial division index, which still requires complex mathematical procedures. Since more large-scale data are used for reception and cause computational bottleneck, we proposed a look-up table method using raster data to improve efficiency.

Dear Reviewer:

Thank you for your suggestion on how to make our research more efficient. Wish you all the best in new year 2021!

Best regards,

Van

Submission Date

27 November 2020

Date of this review

08 Dec 2020 11:09:28

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear authors

This paper considers a relevant issue related to transport planning and logistic control with integrated and comparative approaches. However, I would suggest the following comments to improve the manuscript; 

1. Please revise fig. 1 step 2, as it is not clearly presented. As well as, some minor English changes are required, for example, in Tables 5.
2. I have seen several personal pronouns, it is better to change them to passive voice.
3. Some editing for the manuscript text also requires review, for example, line 441.
4. The comparison between traditional approaches and your method is advisable.

Best regards 

Author Response

Response Reviewer 3

Comments and Suggestions for Authors

 Dear authors

This paper considers a relevant issue related to transport planning and logistic control with integrated and comparative approaches. However, I would suggest the following comments to improve the manuscript; 

1. Please revise fig. 1 step 2, as it is not clearly presented. As well as, some minor English changes are required, for example, in Tables 5.

• Please revise fig. 1 step 2, as it is not clearly presented.

Response 1: Dear Reviewer: We rewrite the whole 4 steps and redraw Figure 1 for the methodology as

bellow (Line 224-249).

Since the return of coordinate data from vehicles is continuous, the fleet dispatching management platform may quickly become overloaded. This paper uses a point-to-polygon method to deal with the high volumes and rates of location GPS data returned by large truck fleets. We also used the same technology to process large fleet location data and found that multiple layer grouping combinations can be used to reduce loading. Figure 1 shows four steps of the process to achieve efficiency improvements with a new multilayer-based algorithm for map matching.

Step 1: Initially, there are two procedures for traffic facility and road data. The traffic facility can be constructed as a polygon. For road data, center lines using vector data are collected from ArcGIS [37]. These centerlines were used to produce buffers, and then the road can be constructed a polygon. For proceeding with multilayer-based map matching, we constructed a large fleet management system based on a structured query language (SQL) database for several segments from road network data to process map matching.

Step 2: Truck fleet data were collected from onboard GPS and plane coordinates of TWD 97 TM 2° obtained for sample testing.

Step 3: The puncture (traditional) method using vector data has more relevant but complex structures that were used to partition and compose the digital map. Thereafter, we used an improved method, adding spatial division index, which still requires complex mathematical procedures. Since more large-scale data are used for reception and cause a computational bottlenecks, we proposed a look-up table method using raster data to improve efficiency.

Step 4: A comparison of the methods based on the two data types was made to provide guidance for

large truck fleet management.

• As well as, some minor English changes are required, for example, in Tables 5.

Dear Reviewer: some minor English changes are corrected in Table 5. We have followed Taiwan’s standard (Column 2: Layer description)

1. I have seen several personal pronouns, it is better to change them to passive voice. (Van)

Response 2: Dear Reviewer:

We have revised some sentence from active voice to passive voice: Line 435, Line 453-454; Line 465-466; Line 514-515

1. Some editing for the manuscript text also requires review, for example, line 441.

Response 3: Dear Reviewer:

We did add the sentences in the previous paragraph, as bellow:

Since the quantity of service stations has fewer data and the position distribution is scattered, service station with a spatial division index may not be efficient (Line 444-445).

Therefore, Line 441 we have deleted.

1. The comparison between traditional approaches and your method is advisable.

Response 4: Dear Reviewer: The traditional well-cited map matching methods are using vector data (puncture method). We proposed a look-up table for comparison. We rewrite the sentences in Step 3 (Line 241-246).

Step 3: The puncture (traditional) method using vector data has more relevant but complex structures that were used to partition and compose the digital map. Thereafter, we used an improved method, adding spatial division index, which still requires complex mathematical procedures. Since more large-scale data are used for reception and cause a computational bottleneck, we proposed a look-up table method using raster data to improve efficiency.

Dear Reviewer:

Thank you for your suggestion on how to make our research more efficient. Wish you all the best in the new year 2021!

Best regards,

Van

Author Response File: Author Response.pdf

Reviewer 4 Report

The paper aims to describe three map-matching methods for large truck fleets and compare their performance. The proposed methods are: puncture method, puncture method with spatial indexs and look-up table on raster data. The 3 proposed methods have been applied in a high workload context on a multi-layer graph and use a linearized coordinate conversion method.
The efficiency results are discussed and analyzed extensively.

Chapters 1 and 2 are broad and discursive by addressing map-matching in different research contexts and showing different context-related topics.

Chapter 3 (Methodology) is unclear.
The first paragraph presents the steps of the methodology which are not mentioned later. In particular, steps 1 and 2, which represent half of the steps, are barely mentioned in the following paragraphs. An exception is the description of the coordinate conversion system, which is described too extensively.
The box in Fig.1 related to the conversion of coordinates in the must be corrected.
The paragraph 3.1 is not clear and Eq. 3 does not find a valid justification. It mentions map-matching but it is a paragraph related only to the multi-layer graph. There is no mention of how to create it.
Paragraph 3.2 on coordinate conversion is superfluous in this context as it is not used in the comparison since it is common to all proposed methods. The citation to [38] could have sufficed.
In section 3.3 AB is described as a polygon but it is a line segment. The formulas adopted are very simple and widely adopted even in non-research contexts. The paragraph talks a lot about the puncture method but neglects the description of the realization of the spatial index and the look-up table. Also there is no technical description of how the methods have been implemented. Is the vector data queried directly on DB or is it in RAM? Are the spatial indexes from GeoDB? Is the lookup table in RAM? In this type of comparison, the implementation techniques are very important in evaluating the results.
Chapter 4 talks about the results but there is an extensive description of the conversion method that is superfluous and should be anticipated in the methodology. It should take less space. The description of the results is very extensive but lacks the description of the implementation choices that can affect the results.

The proposed methods are not innovative and the map-matching is only based on distance, too approximate for moving vehicles. Using direction, speed, shape of links it is possible to give more reliable results. Also the effectiveness between puncture method and look-up table is not analyzed.
The comparison between the methods has predictable results and lacking the description of the implementation choices is incomplete.

Author Response

Response Reviewer 4:

Comments and Suggestions for Authors

The paper aims to describe three map-matching methods for large truck fleets and compare their performance. The proposed methods are the puncture method, the puncture method with spatial indexes, and the look-up table on raster data. The 3 proposed methods have been applied in a high workload context on a multi-layer graph and use a linearized coordinate conversion method.
The efficiency results are discussed and analyzed extensively.

Point 1: Chapters 1 and 2 are broad and discursive by addressing map-matching in different research contexts and showing different context-related topics (Kevin (chapter 2) and Van (Chapter 1) need to discuss).

Response 1: Dear Reviewer:

• In chapter 1, we have referred to citations relevant to map matching, GIS in transportation, map matching algorithms, using cloud computing in urban traffic.

Therefore, in this paper, we use the probe vehicle concept in large truck fleet management which previous researches have not studied yet. Traffic information is collected from the current traffic situation, and the driving time is also provided on each segment of the road to facilitate urgent dispatching.

• In chapter 2: We delete some different context-related sentences (Line 170) and rewrite other sentences in Chapter 2 as bellow.

Developing efficiency on spatial data modeling, road network matching, and visualization for transport in a range of applications [16–18] (Line 115-116).

Luo et al. [22] proposed a hidden Markov model (HMM) map matching method to improve mobile phone positioning efficiency and accuracy and made comparisons using GPS data to yield better performance and demonstrate that such models can help to resolve road matching problems (Line 126-128).

In this paper, Map matching algorithms focus on position error correction and efficiency improvements to show the current positions of the vehicles in general, but large truck fleets are more challenging. Our paper shows most previous studies in Chapter 2.1 on map matching algorithms discuss position accuracy. The efficiency contexts are described or mentioned in the sentences but are not displayed in further studies. Therefore, our paper presented vector-based and raster-based formats and use SIT technology to handle efficiency improvements.

Point 2: Chapter 3 (Methodology) is unclear. (Kevin)

2.1. The first paragraph presents the steps of the methodology which are not mentioned later. In particular, steps 1 and 2, which represent half of the steps, are barely mentioned in the following paragraphs. An exception is the description of the coordinate conversion system, which is described too extensively.

Response 2.1: Dear Reviewer: We add steps 1-4 in sections 3.1-3.3 (Line 251, 273, 296).

2.2. The box in Fig.1 related to the conversion of coordinates in the must be corrected.

Response 2.2: Dear Reviewer: Please see the revised Figure 1. Step 2: Coordinate systems: we use the ShpTrans to transform WGS 84 into TWD 97 TM2 °

2.3. paragraph 3.1 is not clear and Eq. 3 does not find a valid justification. It mentions map-matching but it is a paragraph related only to the multi-layer graph. There is no mention of how to create it.

• Response 2.3: Dear Reviewer: We rewrite the whole 4 steps and redraw Figure 1 for the methodology as

bellow (Line 224-249).

Since the return of coordinate data from vehicles is continuous, the fleet dispatching management platform may quickly become overloaded. This paper uses a point-to-polygon method to deal with the high volumes and rates of location GPS data returned by large truck fleets. We also used the same technology to process large fleet location data and found that multiple layer grouping combinations can be used to reduce loading. Figure 1 shows four steps of the process to achieve efficiency improvements with a new multilayer-based algorithm for map matching.

Step 1: Initially, there are two procedures for traffic facility and road data. The traffic facility can be constructed a polygon. For road data, center lines using vector data are collected from ArcGIS [37]. These centerlines were used to produce buffers, and then the road can be constructed a polygon. For proceeding with multilayer-based map matching, we constructed a large fleet management system based on a structured query language (SQL) database for several segments from road network data to process map matching.

Step 2: Truck fleet data were collected from onboard GPS and plane coordinates of TWD 97 TM 2° obtained for sample testing.

Step 3: The puncture (traditional) method using vector data has more relevant but complex structures that were used to partition and compose the digital map. Thereafter, we used an improved method, adding spatial division index, which still requires complex mathematical procedures. Since more large-scale data are used for reception and cause a computational bottlenecks, we proposed a look-up table method using raster data to improve efficiency.

Step 4: A comparison of the methods based on the two data types was made to provide guidance for

large truck fleet management.

2.4. Paragraph 3.2 on coordinate conversion is superfluous in this context as it is not used in the comparison since it is common to all proposed methods. The citation to [38] could have sufficed.

Response 2.4: The first author had a contribution in [38] (Due to Phonetic transcription, Qin-Yun Mu is Ching-Yun Mu). The coordinate conversion is also related to proceed our proposed multilayer-based algorithm and lookup table method. Therefore, we provide the detailed descriptions.

2.5. In section 3.3 AB is described as a polygon but it is a line segment. The formulas adopted are very simple and widely adopted even in non-research contexts. The paragraph talks a lot about the puncture method but neglects the description of the realization of the spatial index and the look-up table.

Response 2.5: Dear Reviewer: We delete AB on Line 305.

According to our research for handling multiple-layer grouping combinations with the look-up table method, we obtain efficiency improvement without using complicated mathematical procedures. We can use simple formulas for achieving the determination of map positioning immediately.

Previous studies used the puncture method, which is a traditional technology using vector data. We used an improved method, adding spatial division index. We also provide an example (Line 423-427) using Equation (11), (12), and (13) for an explanation. Since more large-scale data are used for reception and cause computational bottlenecks, we proposed a look-up table method using raster data to improve efficiency. Some raster-based information as described in section 2.2 (Line 195-198 and Line 206-217), Spatial Data Structure.

2.6. Also there is no technical description of how the methods have been implemented. Is the vector data queried directly on DB or is it in RAM? Are the spatial indexes from GeoDB? Is the lookup table in RAM? In this type of comparison, the implementation techniques are very important in evaluating the results.

Response 2.6: Dear Reviewer: We would like to explain how the methods have been implemented: This paper uses a point-to-polygon method to deal with the high volumes and rates of location GPS data returned by large truck fleets (Line 225- 226). For adding spatial division index, we used Equation (11), (12), and (13) to obtain index values. The positioning point can be determined on the corresponding block (Line 421, divided into 44 x 80 square blocks). In the look-up table method, we used Digital Differential Analyser (DDA) and Boundary Algebra Filling (BAF) for achieving polygon grids.

The spatial indexes are not from GeoDB. These three methods are implemented based on a SQL database.

For comparison, our proposed multilayer-based method with a look-up table can provide appropriate grouping combinations to handle complicated road networks for managing a large truck fleet.

Point 3: Chapter 4 talks about the results but there is an extensive description of the conversion method that is superfluous and should be anticipated in the methodology. It should take less space. The description of the results is very extensive but lacks the description of the implementation choices that can affect the results .

Response 3: Dear Reviewer: We gave examples to describe how we proceed for conversion method and spatial index method in Chapter 4 (Line 395-427), so the conversion method is not anticipated in the methodology.

We would like to explain how the results have been implemented: Individual computation of each layer is done, and then a combination of all layers together may also be time-consuming. To improve efficiency, we group interchanges and service stations for traffic facilities and the other six layers for roads (Line 450-453). The look-up table method is a competitive solution for using a multilayer-based map matching approach to group eight-segment layers. In the row for the average processing time of every point, the look-up table method provides stable performance using different samples. To evaluate the performance, the multilayer-based map-matching algorithm using a look-up table with raster data provides the best efficiency for large truck fleet dispatching (Line 515-519).

Point 4: The proposed methods are not innovative and the map-matching is only based on distance, too approximate for moving vehicles. Using direction, speed, shape of links it is possible to give more reliable results. 

Response 4: Dear Reviewer: This paper focuses on the discussion of map positioning and efficiency performance, except for accuracy. We proposed a multilayer-based algorithm and look-up table with raster data for improving the performance efficiency under existing software and hardware constraints in large truck fleet management.

Point 5: Also the effectiveness between the puncture method and look-up table is not analyzed.
The comparison between the methods has predictable results and lacking the description of the implementation choices is incomplete. (Kevin)

Response 5: Dear Reviewer: The first author is managing the company for commercial trucks, look-up table provides improvement effectiveness. When the data structure and algorithm procedures are the same, marginal efficiency improvements can be realized by improving the hardware and software. After the hardware and software are upgraded to a certain extent, the marginal efficiency increase will gradually decrease. When determining the investment costs of hardware and software, the actual cost, the market acceptable cost, and the profit should be considered. The best option may not be to upgrade only the hardware and software. The previous study used the puncture (traditional) method with vector data. Applying the puncture method with vector data may yield excess computing loads and delays.

Dear Reviewer:

Thank you for your suggestion on how to make our research more efficient. Wish you all the best in the new year 2021!

Best regards,

Van

Submission Date

27 November 2020

Date of this review

12 Jan 2021 12:28:52

Author Response File: Author Response.pdf

Reviewer 5 Report

It's a good job. Congratulations to the authors.

Author Response

Response: Dear Reviewer: We are sincerely thanks so much for your kind support!

Wish you all the best and Happy new year 2021!

Best regards,

Van

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I acknowledge the commitment of the authors towards the publications of this article. The adjustments made throughout time have clearly improved the quality of the article.

To improve the soundness of the article, the authors may want to consider the following comments:

i) although the introduction has improved in the successive versions of the article, the flow is not clear. The options to address the goal of the article should be kept in the methodology section and the introduction could be more productive if used to provide the context of the research, highlight the gap it is addressing and clearly define the goal of the research;

ii) Although there is a "discussion and conclusion" chapter, there is not a discussion of the findings, only a conclusion. It would be relevant if authors could actually discuss their findings in comparison to results from other researches available;

iii) still related to the previous comment, the overall research is closer to the description of a framework that was idealized by the authors and not so much a framework that emerges from literature. Grounding the decisions in literature is relevant;

iv) the findings may be relevant to the case analysed, but are the findings adequate to other companies? how generalizable are they and which adjustments should be considered to make them useful to other cases?

Author Response

Comments and Suggestions for Authors

I acknowledge the commitment of the authors towards the publications of this article. The adjustments made throughout time have clearly improved the quality of the article.

To improve the soundness of the article, the authors may want to consider the following comments:

Point 1: although the introduction has improved in the successive versions of the article, the flow is not clear. The options to address the goal of the article should be kept in the methodology section and the introduction could be more productive if used to provide the context of the research, highlight the gap it is addressing and clearly define the goal of the research;

Dear Reviewer,

Thank you very much for your comments. Those comments are all valuable and very helpful for revising and improving our paper. We have studied comments carefully and have made correction which we hope meet with approval. The main corrections in the paper and the responses to the comments are as flowing:

Response 1:

Dear Reviewer:

• For the flow of this study:

We add and rewrite the sentences in section 1 (Introduction) as bellow (Line 93-96).

In this paper, we will present a novel method to develop an effective map matching application. With limited resources to upgrade the hardware and software for improving map matching algorithm to implement map positioning, two key factors include time and money (cost) that have to be considered in this study.

• For the point: “Keep the goal of the article in the methodology”

Thanks so much for your suggestions,

We wrote more this sentence in the section 3 (Methodology) as bellow (Line 358-360).

This study aims to use this multilayer-based method with different data structures to effectively achieve improvement for map matching under existing limited software and hardware resources.

The further descriptions are shown in those Line 534-539; and Figure 6 (Section 4).

Point 2:  Although there is a "discussion and conclusion" chapter, there is not a discussion of the findings, only a conclusion. It would be relevant if authors could actually discuss their findings in comparison to results from other researches available;

Response 2: Dear Reviewer: We cite [40] and add descriptions for discussion of the findings.

Based on the conclusion for polygon in raster-based format from [40], we extend the idea to solve problem of large trucks with large-scale positioning data (Line 584-586).

In the fact, our proposed method is efficient, have been applied in practice; and also add the sentences as bellow.

In this study, the smaller truck fleet scenario was not our objective to assess the financial viability, and economies of scale are such that it is probably more feasible economically in large truck fleets, but this was not a consideration in this paper. Therefore, this study only focused on large truck. In the near future, a similar approach can be applied to other regions and countries, and the types of orography can be considered for further research. The multilayer-based map matching approach can also determine fleet locations for further comparisons with different logistics vehicle sizes. (Line 600-606).

Point 3: still related to the previous comment, the overall research is closer to the description of a framework that was idealized by the authors and not so much a framework that emerges from literature. Grounding the decisions in literature is relevant;

Response 3: Dear Reviewer: We cite [40] and add descriptions to support our findings.  

According to the literature in [40], the authors concluded the polygon in raster format using grid method that is more efficient due to few mathematical computations. Therefore, we extend the idea and propose the third approach that involves a look-up table method using a raster-based format (Line 339-341).

Point 4: the findings may be relevant to the case analyzed, but are the findings adequate to other companies? how generalizable are they and which adjustments should be considered to make them useful to other cases?

Response 4:

• Dear Reviewer: Our findings are adequate to other companies without any adjustments, we also have success cases.

The method contributes to large truck fleet management and shows that multilayer-based map matching with a raster-based data structure can improve fleet dispatching. Besides, a multilayer-based map-matching algorithm using the look-up table method does not require complicated mathematical equations or programming languages and can yield the lowest average processing time. When the sample size increases, the look-up table method offers stable performance. With the increase of large truck fleets, our proposed approach can handle the computational bottleneck. Generally, mobile phones have limited computational ability, but installing a multilayer-based map-matching algorithm with a look-up table in the dispatching system can reduce load, and logistics technicians can operate obtain fleet coordinate information via mobile phone (Line 586-594).

Dear Reviewer,

Thank you so much to guide us and we did try our best to improve this research.

Much appreciate your support! Wishing you all the best in 2021!

Sincerely yours,

Van

Submission Date

27 November 2020

Date of this review

31 Jan 2021 16:18:15

Author Response File: Author Response.pdf

Reviewer 4 Report

The changes made to the article are satisfactory and improve understanding of the goals and methodology.

The authors responded a clear and accurate manner to the reviewer's comments by making appropriate improvements to the research.

The reviewer re-evaluated the soundness of the research after the authors' comments, explanations, and improvements.

The latest release of the article appears valid for publication.

Thanks for your work.

Author Response

Response to Reviewer 4

 Comments and Suggestions for Authors

The changes made to the article are satisfactory and improve understanding of the goals and methodology.

The authors responded a clear and accurate manner to the reviewer's comments by making appropriate improvements to the research.

The reviewer re-evaluated the soundness of the research after the authors' comments, explanations, and improvements.

The latest release of the article appears valid for publication.

Thanks for your work.

Response to Reviewer

Dear Reviewer,

Thank you so much to guide us and we did try our best to improve this research.

Much appreciate your support! Wishing you all the best in 2021!

Sincerely yours,

Van.

Submission Date

27 November 2020

Date of this review

26 Jan 2021 10:09:06

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Large truck fleet dispatching is a practical GIS application which raises crucial issues both in logistic system development and design of algorithms to enable the overall dispatching efficiency. This paper investigates the potential in improving map matching process for such purposes. 

However, the reviewer finds the paper fails to identify the algorithmic challenges in above mentioned domain and suffers from poor presentation of the major scientific contribution. 

• in the introduction, the challenge raise by the big data of truck fleet dispatching should clearly introduced, and rational of improving map matching process for such purposes should be discussed.
• in the methodology, the proposed multi-layered method is not clearly discussed
• in the experiment, the improvement in terms of time efficiency of multisample runs seems tedious.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

1) the article would greatly benefit from revision by an English native speaker. The text is not easy to follow and the style required improvement.

2) The introduction could be refined to improve the clarity of the gap the article intends to address. Additionally, substance (references) should be provided to support the statements offered.

3) Methodology stated what will be done in the paper but could be improved by offering the reasons why such options were adopted instead of other options.

4) Internal consistency would improve if the data collection protocols are disclosed.

5) Findings and discussion should be separated into different chapters, allowing more relevance to the real discussion of the results of the research when compared with results from researches from other authors.

6) The discussion section is also quite descriptive rather than analytical. That is, it is very much at the level of describing the results rather than providing the implications of the findings. Currently, the impact/contributions refer to this specific situation, I encourage authors in future iterations of the paper to try to focus more on how this study can contribute to the knowledge in the area (implications, contribution).

7) Findings were produced under specific conditions (options). It would be expected that reference was produced on how those specific conditions influence the findings. How would this approach be economically viable for smaller truck fleet scenarios, is something that could also be addressed.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper deals with multi-layer map matching technique to improve the performance for large truck dispatching problem. This is easy to read and well-explained-written. 

One thing I would like to see more is the implication of this paper. Overall, methodology and result explanation is well-written, and there is no deep discussion about how we apply this method, that is, needs more talks on the contribution of suggested technique with current issues and how we mitigate the issues with this technique. 

In this aspect, there needs more explanation about the current issues for truck fleet dispatching in Related Works session. It will help readers understand about why this work requires.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

1) The authors improved the English writing of the document in a very substantial way. Nonetheless, the document is still wordy and the sequence of ideas offered requires more clarity. These remarks were provided in the list of remarks to the previous version of the document and were not addressed.

2) The article focusses on reporting a project but lacks scientific justification of why the specific options were adopted when alternative solutions are available. If authors include in the methodology a clear sequence of the steps of what was done to achieve the final solution, it would greatly improve the clarity of the methodological sequence used in the research. 

3) A large volume of data is used by the authors. The origin of that data (region and country where it came from) and the type of orography where it comes from should be provided. In the discussion of the findings and in the conclusion of the article this would be revisited as it can greatly influence the ability to generalize the findings.

4) The article still requires clarification of the contribution of the article to the current knowledge in the area (contribution to theory) and its implication to companies (contribution to practice). It is unlikely that truck drivers will directly use the proposed method. Consequently, it would benefit the article if the authors address how the practical contribution of the article is concretized.

Author Response

Please see the attachement

Author Response File: Author Response.pdf