A Collaborative Approach for an Integrated Modeling of Urban Air Transportation Systems

Round 1

Reviewer 1 Report

The authors presented a holistic approach that evaluates the UTM system through taking various relevant subsystems into account and have conduced an in-depth analysis and extensive literature review. Nevertheless I do have few remarks:

• Figure 2: From reading the explanation in the text around figure 2, my understanding is that the two right-hand blocks represent a “Complex system” and a “Chaotic project” respectively.  This difference between system and project should be reflected in Figure 2 by naming the two right hand blocks “Complex system of systems” and “Chaotic project” respectively. The second block can be named “complex system”
• Figure 6: first comment: What about adding disruption management because disruptions are unavoidable see DOI: 1109/TETC.2015.2439633. Second comment: Typically an N2 chart is populated on both sides of the diagonal. Lower half is empty.
• Figure 8: The right hand-side is not visible. Maybe good to remove it because figure 7 already gives a detailed overview. The RCE in the background of figure 7 can be better reduced or removed to not disrupt the quality of the figure.
• Figure 11(a): I understand that the data was collected within a one-week period. Maybe nice to include weather information during that week.
• Figure 3: The percentages on the left do not add up
• Section 3.3 It would be nice if the authors discuss the implications for UTM on urban infrastructure and car traffic. e.g. is it desired to have parking spaces near vertiports, or it will be like an uber service?
• Some minor comments:
  • 2: UAM instead of uam
  • 7: RCE acronym was not introduced before
  • 7: i. a. you mean i.e.?
  • 11: CPACS acronym was not introduced before
  • Figure 2: Reference missing
  • 45: Why therefore? There are also other approaches, not only the SOS. E.g. Agent-Based Modelling
  • 46: Use section instead of chapter
  • 49: differ à differs
  • 63: modal à model
  • 72: MDO acronym not introduced before
  • 75: . [35] à [35].
  • 79: An aircraft is not a complex system
  • 579: und à and
  • 576: a MDO à an MDO
  • Figure 5 shows multiple stakeholders involved in RCE, but no explanation is provided about the role of these stakeholders.
  • 147: usually when you cite someone and then you use by or verb, you associate the name of the author to it. Therefore, Author Name [13] integrated …..

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper deals with a rather innovative topic and at the same time provides an overview of the main elements characterizing a UAM modelling approach, which are the main strengths of the proposed work.

Just some remarks for improving the document, as follows:

All the UAM key concepts should be listed in the first part of the document, in order to identify and arrange in a proper way the main parameters/variables affecting the analyses

According to the holistic approach addressed by the authors, the main research objectives (general and specific) should be better explained to check their pursuing once the main analyses/simulations have been performed.

Regarding table 7, it is suggested to add a legenda (a further column) aimed at showing all cost items in a glance (instead of explaining any single meaning in the text), thus making easier the understandability of proposed formulas.

Some assumptions sound as too simplistic (e.g. the unit charges for ATM and terminal usage are varied in a range between 0.25 and 2; the number of start and parking positions of each vertiport is assumed to be infinite), it could be useful to better justify (or contextualize) them.

The proposed case study covers a peculiar context (3 networks in the Hamburg city), so it would be useful to mention a sort of potential transferability of the main concepts (or criteria) to other urban environments.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

It is an extensive paper about an approach for system of systems modelling for urban air mobility. This is a very relevant topic for the aerospace field and an integrated modelling approach is indeed required to assess the interrelations between the range of aspects as presented in the paper. The paper refers well to relevant literature and it well addresses a broad of UAM issues. Two parts can be distinguished in the paper. Firstly, Sections 2 and 3.1 give a high-level overview over the integrated modelling approach. Secondly, Sections 3.2 to 3.8 present overviews of models and initial results for key elements, such as demand, route networks, flight scheduling, trajectory simulation, and costs.

In general, the first part (Sections 2 and 3.1) is mostly at a conceptual level and parts of it are quite hard to understand, as explained below. Overall, it does not support the reader in getting an overview over the models and their integration. This part should be improved considerably.

• Figure 3: The elements in the left pie chart do not add up to 100%. It is not clear what is meant by “promises of use” being technical, general, etc. Also the pie charts are inconsistent with the reference source [32], e.g. in [32] 49% of the barriers is indicated as technical, while this is not even an element in the right pie chart of Figure 3. Probably best to leave out Figure 3, as it is not a result of the paper and not really needed.
• It is argued that the RCE integration software (Figure 5) is a pillar of the UAM modelling for distributed processing. How this worked for the UAM modelling and how effective this has been is however not shown in the paper. This needs to be improved.
• It is argued that CPACS can be effectively used as a standardized data model (Figure 4b and 4c). It is however not shown in what way CPACS has been adapted / applied for the UAM modelling. This needs to be explained in more detail.
• The notion of “Design Camps” is indicated as a pillar. However, it has not been explained how these are organized and how they can be effective. More concrete terminology should be used than e.g. “…investigate disciplinary correlations within the team” (page 5).
• Figures 7+8-right. The overall architecture of the UAM modelling is hard to read. Better leave out the text RCE to improve readability. Enlarge the figure. Use a larger font for the names of the modules. Explain in detail all modules and their interactions in the text and do not only use some examples in the explanation. The use of colours of the modules is not clear. Also the use of colours is not consistent with that of Figure 8-right. I would suggest to use one colour for the current workflow and another colour for the additional modules in the overall architecture. Then Figure 8-right is no longer needed (it is not readable anyhow).
• Figure 6. The N2-chart for interrelations between the modules is hard to understand. It is not clear what is meant by what is meant by initial, iterative, evaluation, feedback information. Understanding of the figure is hampered by the lack of column names and by having “1” and “i” as symbols. Also the faded colours are not explained. Probably it would be clearer to the reader to explain an improved version of Figure 7 and leave out Figure 6.
• Figure 8-left. It is not readable and has colouring inconsistent with Figure 6. Better leave it out.
• Table 1. This table with flight performance details does not fit in Section 3.1. It could be included in Section 3.6.

The second part (Sections 3.2 to 3.8) presents quite concrete models and results for the Hamburg metropolitan, which are mostly clearly presented and their limitations are well discussed. Also some examples of couplings of the models are discussed. It would be nice though to have some more structured overview of the interconnections between the various models as part of these sections.

• Figure 27. Provide the absolute operating costs in the pie charts and not only the percentages (the percentages are obvious from the pies in the chart anyway).

The last section presents “conclusions and outlook”, which mostly summarizes some key aspects and results of modelling. A discussion of the strengths and weaknesses of the integrated modelling approach is largely missing though and this should be improved. Some relevant topics for the discussion are:

• How effective is the current single data format and what needs to be improved?
• How effective is the coupling between the various modules? For instance, coupling of trajectory simulation and conflict detection requires very high updating rates. Can this be well achieved by RCE?
• What models are now limiting the overall validity of the simulation results and would need to be further improved?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Paper has been further improved by the authors

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

All comments and remarks have been fully accepted, thus improving the document accordingly. The above implies that such a paper, in its revised form, can be accepted.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

I appreciate the way that the authors have addressed the review comments in the paper and explained their reasoning in their letter. As a result, the paper has clearly improved. In particular, the use of RCE has been better substantiated by Sections 2 and 4, the interactions between the model components have been well clarified by the new text in lines 128 to 152, and by the additional text in lines 155 to 162 in combination with Figure 9, the conclusions and outlook has been well improved.

Remaining minor comments that I’d advise the authors to take into consideration are the following.

• The interpretation of the N2-chart in Figure 6 has been improved by the added clarification in the text and by the example in Figure 5. Nevertheless, I advise to further support the reader by adding some definition/explanation of the phases and their data relations, somewhere near line 120.
  • What is the initialization phase (1, orange)?
  • What is the iterative phase, and what is the distinction between a connection (i, green) and a feedback connection (f, red) in the iterative phase?
  • What is the evaluation phase (e, red)?
  • What are data links (faded) and in what ways do they differ from the non-faded connections?
• Figure 7 has been well improved; it is clear and readable now. It is only referred to in line 165; it may be useful to refer to it earlier near line 128.
• Figure 9 is a useful new graph, which not only clarifies the CPACS interface, but also the workflow. Components 1 (Urban Data) and 2 (Airspace Concept) are not explicitly mentioned in this figure. These components could be indicated as part of the Initialization in Figure 9 (or else if more appropriate).
• The readability of Figure 8 has been improved, but I still think that its added value is small. The new Figure 9 presents the current workflow and data interactions much clearer. Of course, it is up to the authors and editor to decide on the figure.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf