Prediction of Compressive Strength and Elastic Modulus for Recycled Aggregate Concrete Based on AutoGluon

Round 1

Reviewer 1 Report

I would be most grateful for invitation to review sustainability-2436125. The title of this paper is “Prediction of Compressive Strength and Elastic Modulus for Recycled Aggregate Concrete Based on AutoGluon”. I consider this manuscript a very interesting review paper. The manuscript suits Sustainability journal scope, and the results are valuables for the recycled aggregate concrete sector. In my opinion the paper worth to be published in Sustainability.

The major comments are listed as flowing:

1. Please provide additional explanation on the basis for the 7:3 division between training data and test data.

2. There is an issue with the statement in lines 378-380, which suggests that WeightedEnsemble performs best in predicting or fitting compressive strength and elastic modulus in experiments.

3. The paper attempts to compare the accuracy of traditional and machine learning methods in calculating compressive strength and elastic modulus, but the traditional method does not include variables such as fly ash, silicon ash, and water reducer for corresponding calculations of compressive strength and elastic modulus.

No

Author Response

Thanks so much for your advice. We have already revised our manuscript. But we may be a little bit confused about the second piece of advice. I would appreciate it if you could explain it again.

Author Response File: Author Response.docx

Reviewer 2 Report

It is an interesting study focusing on AutoGluon application of strength prediction of recycled concrete. The paper is well organized with sufficient data investigations and discussions. Generally, this paper is well-written and can be accepted after some minor revisions:

1)    As the main point is the AutoGluon, some discussion on what are the differences and advantages comparing with traditional AI technics are recommended.

2)    The compressive strength and elastic modulus actually have very close relationship, which are not independent, some analysis between two indexes are suggested.

3)    The conclusion part is suggested to be shorter.

4)  Some recent work on the machine learning of RAC are suggested to be reviewed more sufficiently, such as Journal of Building Engineering 63 (2023): 105570; Journal of Building Engineering 71 (2023) 106508; Construction and Building Materials 370 (2023) 130649.

None

Author Response

Thanks so much for your advice. We have already revised our manuscript. We explained the advantages of AG in the "2.1 part" and shortened the conclusion part. As for the second piece of advice, we put the relationship between compressive strength and elastic modulus in the "4.4" part (line 389).

Author Response File: Author Response.docx

Reviewer 3 Report

 

Recycled concrete gives the option to reduce the total footprint. Presented paper is focused on the prediction models of concrete strength and modulus of elasticity should be a good assistant for engineers in choosing of correct mix design for different design situations. But, the final parameters of concrete with recycled aggregates (not only compressive strength and modulus of elasticity but also shrinkage and creep) depend on the parameters of original aggregates.

Total water consumption is an important parameter but in recycled concrete aggregates, we must differentiate between effective and total water/cement ratio. Depending on the quantity of old concrete on natural aggregates, the water absorption of recycled aggregates can warry and then affect the resulting concrete compressive strength and modulus of elasticity.

-          In appendix A and B, the third columns should by probably cement instead of concrete.

Author Response

Thanks so much for your advice. We have already revised our manuscript and changed the "concrete" to "cement".

Author Response File: Author Response.docx

Reviewer 4 Report

1. Background study, problem statement, points out research gaps, aims/objectives, summary of methods, main results, and novelty of research study are not clearly presented under abstract.

2. Some review study needs more supporting references. Suggest as follow:

(a). “Currently, construction waste materials are mainly dealt with by landfill and accumulation”. Ref. Fuzail Hashmi et al. Green Concrete: An Eco-Friendly Alternative to the OPC Concrete. CONSTRUCTION, 2(2), pp. 93–103. doi.org/10.15282/construction.v2i2.8710.

(b). “The current disposal method is a serious departure from sustainable development goals”. Ref. Omar & Muthusamy. Concrete Industry, Environment Issue, and Green Concrete: A Review. CONSTRUCTION, 1(2), pp. 01–09, doi.org/10.15282/cons.v2i1.7188.

(c). “RAC is a sort of environment-friendly concrete which can be obtained by mixing the waste concrete”. Ref. Nicole Liew et al. Performance of permeable concrete pavement containing recycled aggregate. AIP Conf. Proc. 2688, 040010-1–040010-11; https://doi.org/10.1063/5.0113687.

3. Table 3, 4 and 7. Shows good data but lacking of discussions & analysis. The authors only presenting a data without description of mechanism behind the trend.

4. Figure 5 is not enough discussions.

5. Lengthy conclusions and not reflected to main title and aims of study.

Author Response

Thanks so much for your advice. We have already revised our manuscript. And the discussion is given below each table. we shortened the conclusion part and added relevant references to make our essay more persuasive. Your advice offers us great help and thanks sincerely for it again!

Author Response File: Author Response.docx

Round 2

Reviewer 4 Report

The revised manuscript is acceptable. Thus, accept in current form.