Technical Advances in Aviation Electrification: Enhancing Strategic R&D Investment Analysis through Simulation Decomposition

Round 1

Reviewer 1 Report

I was invited to review the work "Technical advances in aviation electrification: Enhancing strategic R&D investment analysis through simulation decomposition" by the authors M. Kozlova, T. Nykänen, J. S. Yeomans in the journal "sustainability".

The above-mentioned study applies the previous published method "Simulation Decomposition (SimDec) by the same authors to analyze the results of the in this study conducted Monte Carlo simulation for evaluating the impact of the specific energy of batteries and power density of the electric motor in business aviation turboprops. By using the "SimDec" method on the results of the numerical simulation, it was concluded that an increased electric motor power density has a significant impact on the improvement of the flying range if the batteries specific energy is actually high. For myself, it is quite surprising, as I would expect that for higher batteries specific energy the contribution of increasing motor power density on the flying range declines as the main contribution comes from the improvement of the batteries. In my opinion this result is quite remarkable.
From my point of view and with my knowledge, I would accept this study for publication. For the final publication the following minor changes need to be made:

1) Please, add some sentences about the used Monte Carlo model in the section or at least mention, where to find more information about it.

2) On page 5, the unit of the specific energy should be mentioned as energy per mass, instead of "how much energy can be stored in a kilo". For example you can write: "... how much energy can be stored per battery mass..."

3) Please add the power density of the electric motor in Table 4 for the sake of completeness.

4) On page 9, please add space between the value and unit in "500km".

Author Response

We are happy to hear that SimDec was able to uncover and effectively communicate surprising results ?

We are grateful for the reviewer’s effort and time for assessing our manuscript. We address each of the comments one by one.

1) We now add the extra clarification on the model implementation to the end of Section 4.

2) Done.

3) Done.

4) Done.

Reviewer 2 Report

Summary: To support strategic decision-making in an environmental context, this article analyses the effects of technological improvements on the flight range of future all-electric aircraft. The main contribution is the application of Simulation Decomposition together with Monte Carlo simulation to produce graphical representations showing the probabilities of achieving improved electric aircraft flight ranges and their relative dependence on nine different scenarios of technological advancement. The results highlight the importance of battery specific energy and that improvements in motor power density become more beneficial for higher levels of battery specific energy. Improving the electric motor power density alone leads to diminishing flight range benefits and hence returns on investment.

General comments: This article matches very well to the aims and scope of the special issue to which it has been submitted and the text, tables and figures are presented in a well-structured manner. The introduction is engaging and clearly communicates the importance of this research topic by referencing relevant articles on the environmental impact of aviation and its mitigation through electrification.

A) Methods: The methodology was sound and outlined step by step. However, more justification could be provided for the choice of aircraft characteristics and input state ranges used in the simulations. Further details needed to reproduce the analysis could be provided. For example, what software was used? e.g. Matlab? If this information is provided elsewhere and referenced, it would help to state that this is the case and specify exactly where to find this information.

The study assumed some aircraft characteristics that were averaged from the characteristics of typical aircraft listed in Table 2. This led to a simulation of an aircraft which attempted to be representative of the full range of aircraft. In my opinion, a more meaningful approach would have been to select one of the ten aircraft listed in Table 2 which most closely matched the average characteristics and perform the simulation using characteristics of an aircraft that actually exists. Nevertheless, the approach taken in this study is acceptable and still achieves the overall aim, i.e. results that support strategic decision making.

B) Results and Discussion: The study made limited attempt to validate/corroborate the findings and give the reader more confidence in the accuracy of the results. I would recommend checking whether the range predictions are confirmed by other publications that predict the range of electric aircraft having similar characteristics and comment on the agreement/disagreement.

C) Conclusion: “the latter two findings would not have been revealed without the SimDec analysis.” These latter two findings could have been hypothesised from the logic presented earlier in the paper (without the SimDec analysis): higher power density motors enable more battery mass, so more battery mass together with higher specific energy batteries should give greater range benefits, hence conclusion (ii). If no improvements to battery specific energy are made, then improving motor power density yields relatively small benefits associated only with more battery mass, hence conclusion (iii). The SimDec analysis confirmed these hypotheses and provided a quantitative visual format that facilitates decision making by non-specialists in aeronautical engineering. (In other words, aeronautical engineers would have been able to deduce these findings directly from the governing equations/theory without SimDec analysis, but the SimDec analysis could help them communicate better to non-engineers.)

Specific comments

1. Table 1 – Is the caption “Technology-specific assumptions” accurate? The table appears to show data collated from a reference rather than assumptions. In what way are the data in the table “assumptions”?
2. Table 3 – Does the “Motor” column at the far right need to be repeated? The other tables do not repeat the first column. The text “Under development” would fit better without the repeated column.
3. P6, “Batteries at current technological development (LiFePo4, specific power of 2 kW/kg, specific energy of 0.12 kWh/kg) would constitute 7.5-15% of the total mass of an aircraft [16].” Is this statement true for all aircraft?
4. Equation (2) – please define “w_s” and “t” in the text.
5. Equation (4) – where does “0.61” come from?
6. P7, bottom, Table 3 shows that 4.2 kW/kg motors existed from 2017. So why was the transition from the “existing level” to the “under-development level” chosen to be 3.0 kW/kg?
7. Figure 1 – Should the y-axis be labelled ‘probability (%)’ as is done in other papers by the same authors to help readers understand the likelihood of each outcome?
8. p9, para 3 – “300% increased improvement in flight range.” Isn’t the increase from 23 to 69 an increase of 200% relative to 23? Alternatively, the flight range improvement increased to 300% of the original improvement.
9. P9, para 3, “This occurs because the weight saved from using a more efficient motor will provide more benefit if batteries are also more efficient.” The word “efficient” could be misinterpreted as meaning the output energy or power divided by input energy or power, since this is strictly the definition of efficiency. It would be better to be consistent and use the terms that are used throughout the text, i.e. ‘power density’ and ‘specific energy’. (Technically, it is also somewhat inconsistent to refer to the power density of the motors when the term specific energy is used for the batteries. Both cases refer to the energy or power per kilo, so it would be more consistent to use the term ‘specific power’ or ‘gravimetric power density’ of the motors throughout.)
10. P10, Conclusion (i) – “the first observation cannot be considered surprising” – It may help to add the reason why it is not surprising: …because (equation (4) shows that) there is a linear relationship between flight range and battery specific energy.

Author Response

We are grateful for the reviewer’s effort and time for reviewing our manuscript and providing such extensive and relevant commentary to help us to improve our work and make it more rigorous. We address each of the comments, one by one.

A) For the purposes of more easily replicating the study, the additional description of the model implementation is now added at the end of Section 4. The reference with the basic model location is included as well.

We understand the point behind the reviewer’s suggestion to use a particular aircraft instead of an average approximation. However, we remain reluctant to adopt this approach for the following reason. In our model, we are using equations and numeric coefficients (e.g. ratio =0.62 or lift-to-drag ratio = 20) estimated in previous research for an average turboprop, which simplifies the physical reality. Using a particular aircraft design would enable the opportunity of much more precise estimations and complex calculations. But this will simultaneously complicate the computations, unnecessarily, and consequently might damage their transparency and the perception of their generalizability. Therefore, we are glad to hear that the reviewer finds the current approach acceptable.   

Explanations of the choice of numerical thresholds for the variables’ states are now provided. We update the breakdown of the states for the specific power of an electric motor to better reflect information presented in the manuscript data. Although the numeric boundaries between scenarios shift a little bit, all the original conclusions still hold. Therefore, we have also added a comment about the relative unimportance of precise numbers behind the thresholds, since we are looking into general relationships rather than precise scenario boundaries.

B) Indeed, our calculation, of course, has been checked multiple times against the earlier achieved results in the literature [16], when setting up the model. We have provided the validation of our computations at the end of Section 4.1 Computation logic. The resulting range of course, would depend on the multiple factors as explained in [16], but our numbers concur for our average case.

C) We essentially agree with the reviewer (though, post hoc, many analytic findings frequently seem obvious) and have duly corrected that sentence, accordingly, to reflect this point.

However, our aim is to use this as a chance to demonstrate how this analysis is perceived not just by aeronautical engineers, but also by business/operations research experts engaging in the strategic R&D investment analysis process. To reinforce this aspect, this is a quote from another reviewer of this manuscript: “By using the "SimDec" method on the results of the numerical simulation, it was concluded that an increased electric motor power density has a significant impact on the improvement of the flying range if the batteries specific energy is actually high. For myself, it is quite surprising, as I would expect that for higher batteries specific energy the contribution of increasing motor power density on the flying range declines as the main contribution comes from the improvement of the batteries. In my opinion this result is quite remarkable.”

Specific comments

1. Of course, the earlier title was there by mistake, now it is fixed
2. Yes, thank you for that observation, fixed.
3. Indeed, this number is provided for a hybrid aircraft with a battery needed to allow for all-electric take-off and ascent in the referenced paper. However, our calculations are based not on that ratio, but on the equation which has been added to the manuscript.
4. Done.
5. Explained.
6. Yes, the decomposition is updated to reflect that.
7. Here the likelihood of outcomes is not relevant and may even be somewhat misleading, since we do not consider different probabilities of future events, but illustrate the whole range of possibilities.
8. Right, thank you, we fixed the phrasing.
9. Thank you for this important comment. We fixed this particular sentence and substituted 'power density' with the more correct term ‘specific power’ throughout the manuscript. However, it seems that the term ‘power density’ has been incorrectly, but widely, accepted throughout the industry. E.g. Siemens uses it as well to describe their motors. https://www.bbaa.de/fileadmin/user_upload/02-preis/02-02-preistraeger/newsletter-2019/02-2019-09/02_Siemens_Anton.pdf
10. Excellent point, added, thank you.