Multi-State Car-Following Behavior Simulation in a Mixed Traffic Flow for ICVs and MDVs
Round 1
Reviewer 1 Report
The car-following characteristics of mixed traffic flow in different fixed states and with different mixing ratios were simulated and analyzed using the CACC and IDM models in this paper, I think it plays a constructive role in the development of the mixed traffic flow theory, so I recommend to accept after minor revision, but there are still several questions and suggestions.
1. In Fig. 1, with the increase of the mixing ratio, the traffic volume in the mixed car-following flow showed a trend of first decreasing and then increasing，why?
2. There are many existing car-following models of ICV. Why did you choose CACC and IDM?
3. It is recommended to add an overview of each section to the introduction.
4. Strengthen the typesetting of the formula format.
5. It is recommended to change the line types in Figure 1, Figure 2 and Figure 3, because they are so similar that I am difficult to distinguish directly from the Figure.
6. Please add the tools you use in the simulation process and the key steps.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper touches important problem of controlling the traffic of intelligent connected vehicles and manual driving vehicles. Authors propose an adaptive cruise control car-following model and the smart driver model as a base for simulation. Authors include steady, acceleration, and deceleration states in the simulation.
The article is generally correct and written in the right language. The selected elements are new, although the number of simulation parameters is quite primary. Good English, although a native speaker's proofreading may be of benefit. Authors should consider the following corrections to the article:
1. The simulation area did not sufficiently highlight the aspects of automatic/manual driving. Both mods carry out similar activities.
2. The literature review is very poor - no items from the last three years. Was only 2018 full of research? Perhaps it will be beneficial to analyze the movement situation, which is the basis for the simulation (see DOI: 10.5604/01.3001.0014.0206).
3. The mathematical models are written sketchily - no assumptions or limitations are fully reported. What indicators for assessing the quality of the solution were used?
4. The traffic model for autonomous and traditional vehicles is written very laconically.
5. In the introduction, there is no emphasized purpose or structure of the article.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper discusses the car-following behaviors in a mixed traffic flow, i.e., intelligent connected vehicles (ICVs) and manual driving vehicles (MDVs). My main concerns are as follows;
- Lines 18-20 in the abstract - Positive correlation between traffic volume and speed in un-congested situations, and negative correlation between traffic volume and speed in the congested region is not new knowledge. Fundamental diagrams of traffic flow describe the same. It is not clear why this result is highlighted in this paper (in the abstract and conclusions). Does this mean that there are no differences between mixed (ICV and MDV) and MDVs in terms of macroscopic characteristics?
- Two car-following models have been considered in this paper, i.e., CACC and IDM, as described in Sections 3.1 and 3.2. In both models default, parameter values have been used. The values may have been calibrated for different traffic conditions. The authors should provide some discussions to justify the use of the parameters used in previous studies.
- Equation 3 – How those parameters were set? Please provide references (if any) to justify the suggested values.
- Lines 171-172 (“It is assumed that there are only two kinds of car-following behavior in the traffic flow…”) – This statement is not clear. Does this mean both models were used at the same time? Car following situations can be different and the behaviors could also be different depending on such differences. For example, there could be situations where ICV is following an MDV, MDV is following an ICV, ICV is following an ICV, and MDV is following an MDV. Were such differences considered? How the models and parameter values were determined for such different car following situations?
- What is meant by "... acceleration was set as 0.2, 0.25 and 0.3 m/s2..." (lines 213-214). How those values were set? Were those values set for each vehicle? As far as I know, acceleration is an output value of a car following models. (See the equations provided). If the acceleration (or deceleration) was set for a single vehicle, the behavior of the following vehicles will be changed.
- What is meant by the acceleration of the platoon (line 242)? Is the average acceleration of all vehicles? How the variations in acceleration (and deceleration) of a single vehicle were incorporated here?
- Mainly traffic volumes and flows have been presented as results (Fig. 1, 2, 3, and 4). Car-following behavior is rather microscopic. I.e., a car is following another. Why macroscopic characteristics, e.g., change in speed of consecutive vehicles, and platooning vehicles, which are more important, have not been presented or discussed?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The authors made changes and added additional content to the text. The article is now able to be published
Author Response
We would like to thank you for your comments and work on our manuscript.
Reviewer 3 Report
The authors have addressed most of the comments provided by the reviewer. I have a few more comments mainly on Fig. 4.
The lines in Fig. 4 should be in different colors. Currently, they cannot be distinguished.
A reaction time of 1.5 s (according to Table 1) is set in the car-following simulations. However, I cannot clearly see the impact of that reaction time in the results presented in Fig. 4. [An idea about the reaction time can be obtained by observing peaks]. It is slightly visible in the results in Fig. 4(d) and in other figures, the reaction time (of the following vehicles) is not clear. Please check if the reaction time is set appropriately in the simulation models.
Furthermore, the speed variation seems very high implying that the vehicles accelerate and decelerate within short time intervals. Why such sudden acceleration/ deceleration cases were used?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 3
Reviewer 3 Report
The authors have addressed my comments. The manuscript has also been modified accordingly.
