Applying Data Mining Approaches for Analyzing Hazardous Materials Transportation Accidents on Different Types of Roads
Round 1
Reviewer 1 Report
This manuscript used the association rule mining techniques to analyze the hazmat transportation accidents on different road types. The analyses focus on the road transport crashes involving hazmat extracted from the 2016–2020 reported crash data from the Hazardous Chemicals Registration Center of the Ministry of Emergency Management of China. The data driven aspect of this work is quite interesting and informative. However, it also contains many issues that prevent it from being published in the current form. The detailed comments are listed as follows.
- Many works have been done in the literature to investigate the hazmat-related risks, both qualitatively and quantitatively. A lot of risk assessment models have also been developed and applied. It is not obvious why a data-driven method is necessary. To this end, the introduction can be enhanced by adding more discussions of the motivations, especially practical indications, and the literature review should also be extended to provide a state of art survey about the existing risk research. On the basis of the above, the advantages and improvements of the proposed system need to be stated clearly.
- Following the last comment, the main contributions of this work should be properly concluded and clarified.
- Only property damage and casualty are considered in this work. The ecological damage should also be considered.
- The scope of applications and limitations should be discussed.
- The selected factors related to hazmat transportation accidents should be explained. Many factors are related to hazmat transportation accidents. Some of them are considered as the key issues that lead to the accidents. Why this work selects the weather, surface type, time of day, etc.? Moreover, the meaning of the selected factors should also be elaborated. For instance, what is the surface condition, and how this is evaluated?
- Different types of hazardous materials have different degrees of threats to the surroundings, and thus have different effects on the accidents (especially the consequences). However, the type of hazmat is not considered in this work. Please explain.
- Several typos and minor errors are to be amended.
Author Response
Response to Reviewer 1 Comments
Point 1: Many works have been done in the literature to investigate the hazmat-related risks, both qualitatively and quantitatively. A lot of risk assessment models have also been developed and applied. It is not obvious why a data-driven method is necessary. To this end, the introduction can be enhanced by adding more discussions of the motivations, especially practical indications, and the literature review should also be extended to provide a state of art survey about the existing risk research. On the basis of the above, the advantages and improvements of the proposed system need to be stated clearly.
Response 1:
we gratefully appreciate for your valuable comment, we further discussed the motivation for this paper, while expanding the existing literature on the state of the art in risk research to reflect why a data-driven approach is needed. Below are the responses we made.
Road traffic accident history data itself has the characteristics of discrete, multidimensional and fuzzy factor collection. Especially for hazardous chemical road transportation accidents, after the accident, the physical and chemical characteristics of hazardous chemicals themselves will bring about domino effect and the accident characteristics data are relatively large. In addition to the different physical and chemical characteristics of hazardous chemicals themselves, the transport vehicle technology of hazardous chemicals, driver technical requirements are also different from general cargo transportation. Traditional statistical methods are difficult to determine the intertwined relationships with a series of chain reactions.
Among the numerous correlations, it is not easy to use statistical methods to identify the characteristics that have a strong correlation with the severity of accidents. Therefore, it is difficult to accurately predict the severity of accidents, and the relevant prevention and management measures formulated are difficult to be accurately implemented and achieve their expected effects.
Data mining technology is a data analysis method that mines concepts, rules, laws and patterns that are implicit, unknown and potentially valuable for decision making from a large amount of data, and its process is user-oriented and knowledge discovery-oriented. By conducting data mining research on a large amount of historical road traffic accident data, we can analyze the influencing factors of road traffic accidents from multiple perspectives, at multiple levels and in a more comprehensive manner, and reveal the potential laws and characteristics of the interrelated effects of various accident data. In this paper, we mainly use the association knowledge mining and predictive knowledge mining in data mining.
Associative knowledge reflects the degree of dependency between an event and other events, which is expressed in the database in the form of association between structured data structures. Such association forms can be classified as causal association, temporal association, quantitative association and so on. Their common feature is that they are completely unknown and need to be identified by association knowledge mining to find out the hidden association knowledge dependent in the database. Using association mining, the features that have strong association with the occurrence of accidents of different severity on different road types are identified, at which point the identified features will be used as input for the next step of accident severity prediction.
Predictive knowledge is a pattern generated from historical and current data that can predict future trends. The methods used include both the classical regression analysis in statistics and the theories and methods in the field of artificial design via network and machine learning. Based on the scarcity of hazardous chemical accident data, multiple methods are evaluated based on the same training and testing data, and a suitable analysis method is identified for each road type of accident data.
The changes to the manuscript have been marked using the "track changes" function.
Point 2: Following the last comment, the main contributions of this work should be properly concluded and clarified.
Response 2:
We gratefully thanks for the precious time the reviewer spent making constructive remarks. Based on your suggestions, the main contributions of this work have been properly concluded and clarified.
The changes to the manuscript have been marked using the "track changes" function.
Point 3: Only property damage and casualty are considered in this work. The ecological damage should also be considered.
Response 3:
We are very grateful to reviewer for taking your valuable time to provide constructive comments. The reviewer's question about the damage caused to the ecological environment is also an urgent issue in the current collection of hazardous chemical accidents. However, due to the lack of professionals and the absence of a perfect system in the collection of information on hazardous chemical road transport accidents in China, the data obtained so far are only simple descriptions in terms of environmental damage, which are difficult to quantify. Therefore, only property damage and casualties are considered in this paper as a reflection of the severity of accidents. We have two kinds of thoughts about collecting information on the degree of damage caused by accidents to the ecological environment. One is to discuss with experts in the field of environmental engineering to establish a standard system that can quantify the environmental damage. The second is to translate it into property damage. The impact caused by the accident is tracked for a long time to discover the economic loss caused by the environmental damage to the production and life of the surrounding residents. For example, the cost caused by the need to relocate if the environment makes it difficult for the surrounding residents to live.
Point 4:The scope of applications and limitations should be discussed.
Response 4:
We are very grateful to the reviewer for taking your valuable time to provide constructive comments. We apologize for not discussing the scope of applications and limitations. The discussion of application scope and limitations has been added.
The changes to the manuscript have been marked using the "track changes" function.
Point 5:The selected factors related to hazmat transportation accidents should be explained. Many factors are related to hazmat transportation accidents. Some of them are considered as the key issues that lead to the accidents. Why this work selects the weather, surface type, time of day, etc.? Moreover, the meaning of the selected factors should also be elaborated. For instance, what is the surface condition, and how this is evaluated?
Response 5:
Thank you very much for your comments, and we are sorry for the confusion in the selection and interpretation of the factors. The selection of factors in this article is based on the contents covered in the China Road Traffic Accident Information Collection Item List and the Electronic Waybill for Road Transportation of Dangerous Goods. Surface condition indicates the slippery degree of the road surface and is evaluated according to whether there is water, snow or ice on the road surface.
Point 6:Different types of hazardous materials have different degrees of threats to the surroundings, and thus have different effects on the accidents (especially the consequences). However, the type of hazmat is not considered in this work. Please explain.
Response 6:
We gratefully thanks for the precious time the reviewer spent making constructive remarks. Regarding the reasons for not mentioning the types of hazardous materials in the paper, we have made the following explanations. Several reasons are listed for explanation.
(1) It is related to the parameters chosen for determining the set of frequent items and establishing the association.
(2) It is related to the fact that in the discussion section, only the top 10 rules with the highest lift values related to the severity of the accident were selected, without covering the full range of correlations.
When we read other articles that study the factors influencing accidents in the road transport of dangerous goods, we also find that their conclusions do not cover the type of dangerous goods. For example, Essential Safety Factors for the Transport of Dangerous Goods by Road: A Case Study of Lithuania. The study identified three group factors that have the greatest impact on the safe transport of dangerous goods by road. The outcomes of the research presented in the paper show that the main factors of Group I have the greatest impact on the likelihood of an accident during dangerous goods transportation by road transport; they include incorrect loading of cargo, driver fatigue, vehicle condition, and weather and road surface conditions. The most important technical/technological factors of Group II are correct loading or unloading of the cargo, tightness of the vehicle semi-trailer/container/tank, and the technical condition of the vehicle. The most important organizational factors in Group III are the risks associated with the carriage of goods, the choice of route, and communication with the emergency services.
Although the correlation between the type of hazardous chemicals and the severity of the accident is not directly mentioned in our paper, the type of hazardous materials transport vehicle, hazardous materials loading equipment, and hazardous materials driver's qualification all have a correlation of some strength with the severity of the accident. The special nature of dangerous goods transportation, in addition to the physical and chemical characteristics of dangerous goods themselves, their transportation, reloading equipment and requirements for drivers are also different from other goods transportation.
Point 7:Several types and minor errors are to be amended.
Response 7:
We are very grateful to the reviewer for taking your valuable time to provide constructive comments. We have identified a native English-speaking editor to assist us in preparing the manuscript. Using the track changes mode in MS Word, The changes to the manuscript have been marked using the "track changes" function.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear ,
the paper is well organized, the authors dealt with a fairly current topic, in a clear and simple way, I think it is appropriate to point out the following observations: -The abastract is well written, but it must be shorter. -increase the introductory part by inserting international references. --Improve Figure 1 - decrease the discussion section it must be more concise, -increase the section of the conclusions -the English language is good. increase the bibliography. bibliographic suggestions:
Safety evaluation of flower roundabout considering autonomous vehicles operation
Severino, A., Pappalardo, G., Curto, S., Trubia, S., Olayode, I.O. ,
Author Response
Point 1: -The abastract is well written, but it must be shorter.
Response 1:
We gratefully thanks for the precious time the reviewer spent making constructive remarks. We have streamlined the abstract section based on the reviewers' suggestions.The changes to the manuscript have been marked using the "track changes" function.
Point 2: -Increase the introductory part by inserting international references
Response 2:
We gratefully appreciate for your valuable comment. We have studied more international literature based on your suggestions, and we are very grateful to the reviewers for providing us with references, which is very rewarding. The changes to the manuscript have been marked using the "track changes" function.
Point 3: -Improve Figure 1.
Response 3:
We gratefully appreciate for your valuable comment, and we have enhanced Figure 1, from dpi 150, to dpi 300.
Point 4: -Decrease the discussion section it must be more concise
Response 4:
We gratefully thanks for the precious time the reviewer spent making constructive remarks. We have carefully considered and discussed the discussion part of the article, removed some obvious function introductions and repetitive explanations of the results, and only streamlined the rest of the content in terms of language. For other specific contents, no substantive changes were made.
The changes to the manuscript have been marked using the "track changes" function.
Point 5: -Increase the section of the conclusions
Response 5:
We gratefully thanks for the precious time the reviewer spent making constructive remarks.Based on the reviewers' suggestions, additions and changes were made to the conclusion section, and the changes to the manuscript have been marked using the "track changes" function.
Point 6: -Increase the bibliography. bibliographic suggestions:
Safety evaluation of flower roundabout considering autonomous vehicles operation
Severino, A., Pappalardo, G., Curto, S., Trubia, S., Olayode, I.O. Sustainability (Switzerland), 2021, 13(18), 10120
Severino, A.; Pappalardo, G.; Curto, S.; Trubia, S.; Olayode, I.O. Safety Evaluation of Flower Roundabout Considering Autonomous Vehicles Operation. Sustainability 2021, 13, 10120. https://doi.org/10.3390/su131810120
Response 6:
We gratefully appreciate for your valuable comment, and we have carefully studied the article recommended by the reviewer, which are very rewarding. We have added it to our bibliography list，the changes to the manuscript have been marked using the "track changes" function.
Author Response File: Author Response.docx
Reviewer 3 Report
The paper brings a relatively new view on the risk assessment of road accidents involving hazmat (hazardous materials) transportation. The paper used association rule mining (ARM) techniques to analyze Hazmat transport accidents occurring on different road types, using road types as classification criteria.The research results of hazmat transportation also provide support for the decision-making process in regard to hazmat road transportation risk management.The paper should compare the differences with previous work more than just the application of the techniques.
Author Response
Point 1:The paper brings a relatively new view on the risk assessment of road accidents involving hazmat (hazardous materials) transportation. The paper used association rule mining (ARM) techniques to analyze Hazmat transport accidents occurring on different road types, using road types as classification criteria.The research results of hazmat transportation also provide support for the decision-making process in regard to hazmat road transportation risk management.The paper should compare the differences with previous work more than just the application of the techniques.
Response1:
We are grateful to the reviser for the comments, and with your suggestion, we have added contents to the introduction to reflect the differences with previous work，and the changes to the manuscript have been marked using the "track changes" function.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The authors have properly addressed my major concerns. Only one minor additional suggestion is to adjust the font style for some notation (italic). For example, C1-Ck etc. on page 2, as well as in Figure 1.
Author Response
Reviewer#1, Concern # 1:
The authors have properly addressed my major concerns. Only one minor additional suggestion is to adjust the font style for some notation (italic). For example, C1-Ck etc. on page 2, as well as in Figure 1.
Author response:
We gratefully thanks for the precious time the reviewer spent making constructive remarks. Based on your suggestion, we have adjusted the font of the notations to italic. We have also changed C1-Ck in Figure 1 to be italicized as well.
The changes to the manuscript have been marked using the "track changes" function.
Author Response File: Author Response.docx
Reviewer 2 Report
the authors did a great job. the requested changes have been made. the manuscript is scientifically mature. you can proceed with the publication. best wishes to the authors for future works
Author Response
Reviewer#2, Concern # 1:
The authors did a great job. the requested changes have been made. the manuscript is scientifically mature. you can proceed with the publication. best wishes to the authors for future works.
Author response:
Thank you very much for the reviewer's approval. We will continue to work hard.
