Identification of Machine Learning Relevant Energy and Resource Manufacturing Efficiency Levers
Round 1
Reviewer 1 Report
1. Several claims about Machine Learning (ML) seem to be inaccurate and should be backed up by appropriate citations. For example, lines 36-38: [Non-ML approaches] "are differentiated from ML approaches in that they are better suited when the main objective is to understand and model processes, or when processes with limited parameters are to be analyzed."
First, ML does model processes; training a linear regression on the housing price prediction task models the process of mapping the number of bedrooms, house size, date of construction, and others, to the selling price of a given house. The learned weights can also indicate how much a feature affects the predicted price. The claim that non-ML approaches are generally better seems arbitrary.
Second, having a limited number of features does not prevent the use of ML, nor does it mean the result will necessarily be worse than a non-ML approach. This depends on how much data you have, how difficult the specific problem is, etc.
2. "High variety data: Problems and data sets with high dimensionality; i.e. the data to be analyzed has high variety." Are the authors referring to high variance, rather than variety? Data variety usually refers to having multiple / heterogeneous sources from which the data is collected, so I'm inclined to think the authors mean "variance." In that case, it is important to note that high dimensionality does not necessarily equal high variance.
3. "Difficult to capture data: When data cannot feasibly be captured with conventional sensors; i.e. when process can only be observed visually or acoustically" Image / audio sensors are fairly conventional and have been around for decades. Did the authors mean "difficult to interpret?"
4. "The analysis was based on literature and experience of the authors. For each situation that is relevant, a point was allocated to the lever. To avoid an overly long paper, the justification for each point is not included in the paper." These two statements heavily undermine the scientific soundness of this paper.
Author Response
Please see attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
Identification of Machine Learning Relevant Energy & 1 Resource Manufacturing Efficiency Lever” article contributes to the field but needs substantial improvement.
1. Abstract nee improvement in terms of after results and method
2. Authors must include the literature section starting with aspects of the sustainability for the organizations and then lick with environment I have few suggestions in this regards to cover the aspects of the sustainability. “An Analysis of Circular Economy Deployment in Developing Nations’ Manufacturing Sector: A Systematic State-of-the-Art Review “ “Green Lean Six Sigma for improving manufacturing sustainability: Framework development and validation” “Green lean six sigma sustainability–oriented project selection and implementation framework for manufacturing industry” “Exploration and investigation of green lean six sigma adoption barriers for manufacturing sustainability” This will enhance the aspects of the sustainability “Green lean six sigma journey: Conceptualization and realization”
3. Authors presented results in the nice manner but more elaboration on the discussion part is needed.
4. There is a slew typo and grammatical error in the manuscript. Please fix the same.
5. Improve the referencing of the paper. There are some missing references.
6. Compare the results of your study with previous studies of the same nature.
7. Conclusion must represent after effect of the study.
8. Practical Implication must be improved.
Author Response
Please see attachment
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Authors have made all required changes in the article. Article is ready for publication
