Setting Thresholds to Define Indifferences and Preferences in PROMETHEE for Life Cycle Sustainability Assessment of European Hydrogen Production

Round 1

Reviewer 1 Report

The paper addresses the issue of using PROMETHEE method for interpreting LCSA results for hydrogen production. The novelty of the approach is directly stated within the title and is focused on adopting PROMETHEE for this specific use and processing LCSA analysis and results to fits its possible extension.

The paper is well written, referenced to actual and significant literature and has objectively important overall merit. The text has appropriate structure and covers it well with the content. The methodology or the paper is proper one for its objectives. The methodology used for research is very complex though but the supplementary material provides appropriate explanations and examples for its better understanding. The paper could significantly contribute to the development of MCDA toolset with regard to the interpretation and uncertainty of LCSA results handling.

The only drawback of the paper is a complete lack of discussion of the results with existing literature. Therefore, the recommendation is to elaborate the discussion section in the paper. The results of the analysis should be projected versus the results of similar studies. Similarity should be explored in a sense of MCDA methods and its possible use with life cycle based results and also in a sense of hydrogen production LCSA. Additionally, since the paper is focused on uncertainty of LCSA results handling it would be advisable to address this issue as well. Addressing these remarks would certainly help in improving academic soundness of the paper and strength of its arguments.

Author Response

It has turned out that the discussion of MCDA methods is hardly possible as they rely quite strongly on the system investigated. Therefore we chose the approach of comparing our results of specific thresholds to general default thresholds, which represent a common literature average. However, to follow your advice of a more elaborate comparison, we added a more detailed description of other approaches in the first part of the article. This underlines the non-comparability of the approaches.

Reviewer 2 Report

Dear authors,

first of all: thank you the interesting read! This is a very exciting, while not very novel, topic and a well prepared manuscript.

Please consider the follwing recommendations for improvement:

Line 38: please also add criticism on hydrogen here, such as the use of fossil based energy for its production or the low energy efficiency of the process

Line 298: Please explain why you use equal weighting. Within LCA, there are recommended weighting factors (e.g. https://ec.europa.eu/environment/eussd/smgp/documents/2018_JRC_Weighting_EF.pdf). Furthermore, there are methods for calculating weights in MCDA objectively (such as Shannons Entropy or CRITIC).

Line 332: What are the sources for the parameters such as life time, operation time, etc.?

In general:

The assumptions made for the LCA have to be expanded upon substantially. E.g.: What allocation rules did you apply? Which system boundaries (Fig.2 can not be considered as a description of your system boundaries)? Cut of rules? End of life assumptions? What electricity mixes were used (national production or consumption grid mixes, or a supplier or product mix, if yes: what is the mix?)

Additionally, you could add further sources of authors discussing thresholds and uncertainties in MCDA for LCSA

Author Response

Line 38: please also add criticism on hydrogen here, such as the use of fossil based energy for its production or the low energy efficiency of the process

We added also a sentence on the downside of hydrogen.

Line 298: Please explain why you use equal weighting. Within LCA, there are recommended weighting factors (e.g. https://ec.europa.eu/environment/eussd/smgp/documents/2018_JRC_Weighting_EF.pdf). Furthermore, there are methods for calculating weights in MCDA objectively (such as Shannons Entropy or CRITIC).

Discussing the choice of weighting factors might fill another article. Here, we want to reduce complexity simple in this regard and focus on the thresholds. However, we added some sentences addressing this question and giving some references with examples where other weighting approaches have been applied.

Line 332: What are the sources for the parameters such as life time, operation time, etc.?

In general:

The assumptions made for the LCA have to be expanded upon substantially. E.g.: What allocation rules did you apply? Which system boundaries (Fig.2 can not be considered as a description of your system boundaries)? Cut of rules? End of life assumptions? What electricity mixes were used (national production or consumption grid mixes, or a supplier or product mix, if yes: what is the mix?)

We added more information about the case study in the paper. However, the technical system and the LCA is taken without change from the paper Koj et al. 2017 as referenced in the paper. Thus we did not go into great detail with, e.g. the system boundaries.

Additionally, you could add further sources of authors discussing thresholds and uncertainties in MCDA for LCSA

We included in this paper some literature with regard to thresholds, PROMETHEE and LC(S)A as well as thresholds, uncertainties and LC(S)A.