the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Nov 2022
Robust 4D Climate Optimal Flight Planning in Structured Airspace using Parallelized Simulation on GPUs: ROOST V1.0
Abstract. The climate impact of the non-CO2 emissions, being responsible for two-thirds of aviation radiative forcing, highly depends on the atmospheric chemistry and weather conditions. Hence, by planning aircraft trajectories to reroute areas where the non-CO2 climate impacts are strongly enhanced, called climate-sensitive regions, there is a potential to reduce aviation induced non-CO2 climate effects. Weather forecast is inevitably uncertain, which can lead to unreliable determination of climate-sensitive regions and aircraft dynamical behavior and, consequently, inefficient trajectories. In this study, we propose robust climate optimal aircraft trajectory planning within the currently structured airspace considering uncertainties in the standard weather forecasts. The ensemble prediction system is employed to characterize uncertainty in the weather forecast, and climate-sensitive regions are quantified using the prototype algorithmic climate change functions. As the optimization problem is constrained by the structure of airspace, it is associated with hybrid decision spaces. To account for discrete and continuous decision variables in an integrated and more efficient manner, the optimization is conducted on the space of probability distributions defined over flight plans instead of directly searching for the optimal profile. A heuristic algorithm based on the augmented random search is employed and implemented on graphics processing units to solve the proposed stochastic opti- mization computationally fast. The effectiveness of our proposed strategy to plan robust climate optimal trajectories within the structured airspace is analyzed through two scenarios: a scenario with large contrails’ climate impact and a scenario with no formation of persistent contrails. It is shown that, for a night-time flight from Frankfurt to Kyiv, a 55 % reduction in climate impact can be achieved at the expense of a 4 % increase in cost.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(27655 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(27655 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-1010', Anonymous Referee #1, 24 Feb 2023
Review of Simorgh et al.
Simorgh et al. present ROOST v1.0, an open-source Python library for 4D climate optical flight planning in structured airspace with the capability for using GPUs for efficient augmented random search. The authors introduce robust planning through the consideration of uncertainties in standard weather forecasts using an ensemble prediction system. The manuscript presents evaluation of the new model software through the planning of a flight from Frankfurt to Kyiv using a Airbus A320-214 on two different departure times representing scenarios with and without formation of persistent contrails, and demonstrates a climate impact reduction of 15-55% corresponding to 0.8-4% increase in operating cost. The work is novel and represents a significant new model software development, especially its consideration of sources of uncertainty, and is thus well fit for the scope of Geoscientific Model Development. I have minor comments regarding the manuscript before recommending it for publication.
Major comments:
1. One of the main contributions in the study (L89-L92) is the determination of optimized trajectory with fast computational efficiency. Could the authors elaborate on this fast computational efficiency, e.g., (1) how fast does the model run for a given scenario for prediction, (2) considering the use of GPUs, the GPU (particularly memory) requirements for the model, and (3) the data requirements (e.g., EPS forecast data) for running the model? If the computational efficiency is notable compared to other studies, it will be useful to include comparisons to prior work as well, as flight planning is a time sensitive operational task and would greatly benefit from improved computational efficiency if the improvements are significant compared to prior work or those currently used in the aviation industry. In this case, a brief description of how GPUs are used in the work and specific optimizations for future readers' reference will be very useful as well.2. A specific note on the code reproducibility. Because ROOST requires the BADA license for representation of aircraft aerodynamics, the code provided cannot be evaluated as it is incomplete. I understand the authors are not totally in control of this but it would be reassuring for the open-source nature of the software to include a paragraph on potential future implementations of other open-source aircraft performance models within ROOST, and if ROOST has the capability/interfaces for it.
Specific comments:
1. L79: "These studies suffer mainly from computational perspectives and some restrictive assumptions (see Simorgh et al...)" could you briefly include some examples of these restrictive assumptions? Also, it is unclear what is being referred to as "computational perspectives".
2. Page 4, Table 1 - I suggest including "This work" for easy comparison.
3. L98: "optimized trajectory ia tracked as determined" is unclear. Do you mean the optimized trajectory is deterministic but in fact takes into account the uncertainty of the weather forecasts? Please clarify.
4. L190: How are \Psi_CST and \Psi_CLM selected?
5. L210: "aCCFs estimate ... computationally in real-time". I would also point out a particular advantage that it does not require external datasets which helps with compute requirements.
6. The use of altitude and pressure in figures could be more consistent. e.g., Figure 2 uses 250 hPa which is standard for science but later results e.g., Figure 10 use FL360, FL340, etc. which is standard for the aviation industry. To help readers, it may be useful to add estimate of altitude in the Figure 2 legend (250 hPa is approx. FL340), and vice-versa in other figures to help context.Technical corrections:
- L227: "nigh-time" -> "night-time"
- Figure 7: "expected perfromace" -> "expected *performance*"Citation: https://doi.org/10.5194/egusphere-2022-1010-RC1 - AC1: 'Reply on RC1', Abolfazl Simorgh, 30 Mar 2023
- AC4: 'Reply on RC1', Abolfazl Simorgh, 31 Mar 2023
-
RC2: 'Comment on egusphere-2022-1010', Anonymous Referee #2, 02 Mar 2023
The paper by Simorgh and colleagues describes a python library for the climate-optimal planning of flight trajectories within the structured airspace taking uncertainties in weather forecasts into account. The library is designed for parallel simulation on GPUs. The cost function of the optimization problem considers operational costs and climate impacts of aircraft emissions, whereas both factors can be weighted individually. From what is written in the introduction, the presented work seems to represent an important and novel contribution in the field of optimal flight planning.
Overall the paper is relatively clearly written and the performance of the tool is nicely demonstrated by two examples, a night-time flight from Frankfurt to Kyiv during summer and a day-time flight on the same route during winter. My major point of criticism is related to the length of the paper, which hinders readability. Section 2 starts with a formulation of the deterministic climate-optimal flight planning problem, followed by the description of the aircraft dynamical model and the cost function, which is to be minimized in the optimization problem. This subsection (2.1.2) includes a rather detailed description on how the climate impact of aircraft emission is determined. If I understand this correctly, this part has already been published elsewhere. The paper continues with a section on uncertainties in weather forecasts and how optimal flight planning problem has to be reformulated taking these uncertainties into account. This means that several equations occur twice in the paper, once with and once without uncertainty parameters. In my view this is a bit confusing and the reader might easily lose the thread. Maybe the authors find a more concise way to present their method, e.g. by moving some of the equations into the appendix. Also, a short overview/schematic of the approach at the beginning of Section 2 might help the reader to better understand the individual parts and how they are connected. After some minor modifications (for details see attached pdf) I recommend the manuscript for publication in GMD.
- AC2: 'Reply on RC2', Abolfazl Simorgh, 31 Mar 2023
- AC3: 'Reply on RC2', Abolfazl Simorgh, 31 Mar 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-1010', Anonymous Referee #1, 24 Feb 2023
Review of Simorgh et al.
Simorgh et al. present ROOST v1.0, an open-source Python library for 4D climate optical flight planning in structured airspace with the capability for using GPUs for efficient augmented random search. The authors introduce robust planning through the consideration of uncertainties in standard weather forecasts using an ensemble prediction system. The manuscript presents evaluation of the new model software through the planning of a flight from Frankfurt to Kyiv using a Airbus A320-214 on two different departure times representing scenarios with and without formation of persistent contrails, and demonstrates a climate impact reduction of 15-55% corresponding to 0.8-4% increase in operating cost. The work is novel and represents a significant new model software development, especially its consideration of sources of uncertainty, and is thus well fit for the scope of Geoscientific Model Development. I have minor comments regarding the manuscript before recommending it for publication.
Major comments:
1. One of the main contributions in the study (L89-L92) is the determination of optimized trajectory with fast computational efficiency. Could the authors elaborate on this fast computational efficiency, e.g., (1) how fast does the model run for a given scenario for prediction, (2) considering the use of GPUs, the GPU (particularly memory) requirements for the model, and (3) the data requirements (e.g., EPS forecast data) for running the model? If the computational efficiency is notable compared to other studies, it will be useful to include comparisons to prior work as well, as flight planning is a time sensitive operational task and would greatly benefit from improved computational efficiency if the improvements are significant compared to prior work or those currently used in the aviation industry. In this case, a brief description of how GPUs are used in the work and specific optimizations for future readers' reference will be very useful as well.2. A specific note on the code reproducibility. Because ROOST requires the BADA license for representation of aircraft aerodynamics, the code provided cannot be evaluated as it is incomplete. I understand the authors are not totally in control of this but it would be reassuring for the open-source nature of the software to include a paragraph on potential future implementations of other open-source aircraft performance models within ROOST, and if ROOST has the capability/interfaces for it.
Specific comments:
1. L79: "These studies suffer mainly from computational perspectives and some restrictive assumptions (see Simorgh et al...)" could you briefly include some examples of these restrictive assumptions? Also, it is unclear what is being referred to as "computational perspectives".
2. Page 4, Table 1 - I suggest including "This work" for easy comparison.
3. L98: "optimized trajectory ia tracked as determined" is unclear. Do you mean the optimized trajectory is deterministic but in fact takes into account the uncertainty of the weather forecasts? Please clarify.
4. L190: How are \Psi_CST and \Psi_CLM selected?
5. L210: "aCCFs estimate ... computationally in real-time". I would also point out a particular advantage that it does not require external datasets which helps with compute requirements.
6. The use of altitude and pressure in figures could be more consistent. e.g., Figure 2 uses 250 hPa which is standard for science but later results e.g., Figure 10 use FL360, FL340, etc. which is standard for the aviation industry. To help readers, it may be useful to add estimate of altitude in the Figure 2 legend (250 hPa is approx. FL340), and vice-versa in other figures to help context.Technical corrections:
- L227: "nigh-time" -> "night-time"
- Figure 7: "expected perfromace" -> "expected *performance*"Citation: https://doi.org/10.5194/egusphere-2022-1010-RC1 - AC1: 'Reply on RC1', Abolfazl Simorgh, 30 Mar 2023
- AC4: 'Reply on RC1', Abolfazl Simorgh, 31 Mar 2023
-
RC2: 'Comment on egusphere-2022-1010', Anonymous Referee #2, 02 Mar 2023
The paper by Simorgh and colleagues describes a python library for the climate-optimal planning of flight trajectories within the structured airspace taking uncertainties in weather forecasts into account. The library is designed for parallel simulation on GPUs. The cost function of the optimization problem considers operational costs and climate impacts of aircraft emissions, whereas both factors can be weighted individually. From what is written in the introduction, the presented work seems to represent an important and novel contribution in the field of optimal flight planning.
Overall the paper is relatively clearly written and the performance of the tool is nicely demonstrated by two examples, a night-time flight from Frankfurt to Kyiv during summer and a day-time flight on the same route during winter. My major point of criticism is related to the length of the paper, which hinders readability. Section 2 starts with a formulation of the deterministic climate-optimal flight planning problem, followed by the description of the aircraft dynamical model and the cost function, which is to be minimized in the optimization problem. This subsection (2.1.2) includes a rather detailed description on how the climate impact of aircraft emission is determined. If I understand this correctly, this part has already been published elsewhere. The paper continues with a section on uncertainties in weather forecasts and how optimal flight planning problem has to be reformulated taking these uncertainties into account. This means that several equations occur twice in the paper, once with and once without uncertainty parameters. In my view this is a bit confusing and the reader might easily lose the thread. Maybe the authors find a more concise way to present their method, e.g. by moving some of the equations into the appendix. Also, a short overview/schematic of the approach at the beginning of Section 2 might help the reader to better understand the individual parts and how they are connected. After some minor modifications (for details see attached pdf) I recommend the manuscript for publication in GMD.
- AC2: 'Reply on RC2', Abolfazl Simorgh, 31 Mar 2023
- AC3: 'Reply on RC2', Abolfazl Simorgh, 31 Mar 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 488 | 178 | 21 | 687 | 12 | 4 |
- HTML: 488
- PDF: 178
- XML: 21
- Total: 687
- BibTeX: 12
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
1 citations as recorded by crossref.
Abolfazl Simorgh
Manuel Soler
Daniel González-Arribas
Florian Linke
Benjamin Lührs
Maximilian M. Meuser
Simone Dietmüller
Sigrun Matthes
Hiroshi Yamashita
Feijia Yin
Federica Castino
Volker Grewe
Sabine Baumann
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(27655 KB) - Metadata XML