Code Calibration of the Eurocodes

Round 1

Reviewer 1 Report

This paper proposes a code calibration for the Eurocodes. In general, the paper is well-written and the idea that the Author pretends to transmit is solid. However, the Author must validate and/or demonstrate that such a calibration is necessary based on the current state of the art. In this way, the Author must increase the technical background of the paper to justify that a code calibration is necessary. In other words, the literature review must be increased to demonstrate that the calibration is valid. In addition, the following revision must be performed to the paper.

1. In Section 1.1, the discussion must be expanded to explain more in detail why building codes calibrated with accuracy level II decrease the target reliability and cause a load reduction. Please include some references where this have been documented.
2. In Section 1.2, the Authors is documenting several weaknesses of current Eurocodes. However, only two citations (13 and 14) are included in such a discussion to justify them. Please include more technical information reported preferably on international journals about such weaknesses of the current Eurocodes.
3. In Section 2.4, why variable load is assumed to have a Gumbel distribution? Is this based on other studies. Please justify it.
4. Section 2.6 is hard to follow. Please consider expanding the discussion about Uncertainty.
5. The is a typo y the title of Section 2.7. Please change “6. The basic equtions” to “2.7. The basic equations”.
6. The title of Fig. 1 is very long, please short it down. Please consider including labels inside Fig. 1 for the different types of lines included in the illustration.
7. The title of Fig. 2 is very long, please short it down. Please consider including labels inside Fig. 2 for the different types of lines included in the illustration.
8. In some part of the manuscript, the Author must include the limitations of the work.
9. The title of Fig. 3 is very long, please short it down. Please consider including labels inside Fig. 3 for the different types of lines included in the illustration.
10. Please change conclusions to bullet points.

Author Response

Reviewer 1

Thank you for the review.

The amendment of the Eurocodes is necessary to remove the excess safety and material waste for loads V_M = 0.2 explained in sections 3.2 and 4. Simultaneously, simpler codes with the same load factors g_G = g_Q are obtained. Other advantages are achieved, too.

Two articles are added to the reference list addressing the code calibration. The author did not find an article which would propose a similar amendment ‑ variable characteristic values ‑ as presented here.

1. The codes with the accuracy level II are based on the independent load combination and 5-year variable loads which result in about 10% too low safety factors explained by the author earlier [17-20].
2. The author did not find articles explaining explicit deficiencies of the Eurocodes. Many structural designers consider the codes complicated which is mainly due to the load combination.
3. All relevant earlier research on structural codes e.g. [1,5-16] assume that the variable load distribution is Gumbel.
4. The uncertainty calculation is unclear in the structural probability theory. Various calculation methods are presented. No consensus exists. In the author’s opinion the uncertainty should be omitted as no relevant data for the uncertainty parameters exist.
5. A typo on line 226 is corrected.
6. The title of Fig 1 is made shorter. The author uses MATHCAD 15 program which has a poor graphical output. The quality of Figures is improved.
7. The title of Fig 2 is made shorter.
8. A paragraph including the limitation of the article is added in the beginning.
9. The title of Fig 3 is made shorter.
10. The conclusion is amended to include bullet points.

Reviewer 2 Report

• The journal articles in the reference list are limited, I recommend adding some journal articles that talk about this subject.
• P1; L27: I recommend adding a section to present the article layout.
• P2; L58: The terminology must be defined when mentioned for the first time in the article (Gama_G, Gama_Q ).
• P2; L58: Could the author explain how not equal load factors make the codes prone to design error.
• P2; L72: meaning of timber VM=0.2 and why it is considered as ideal material.
• P3; L100: meaning of Gama_L
• P3; L112: Please, add a short definition to Gumbel distribution.
• P3; L138: Could the author explain "the characteristic load of the variable load is set as adaptable "
• P3; L136: What is the difference between the dependent load combination and the independent load combination?
• P4; L181: It is good to briefly define these kinds of statistical distribution or add some figures to present them to the unspecialized reader.
• References to many equations and defined functions are missing.
• P6, L 237: Could you please give a short definition to Monte-Carlo simulation.
• P6; L247: Does the author mean the codes or equations in the codes fulfill the requirements of the code?
• P6; L249: I recommend adding an appendix that contains the calculations or sample of calculations in order to enable the interested reader to follow the calculations and maybe verify them.
• P6; L251: How the author has selected these values for the coefficient of variations.
• Add legend to figure 1,2 & 3
• P9; L402: If I understood correctly, the author proposed equations for load combinations while the codes are covering many aspects of the engineering knowledge.
• P9; L407: But the SLS is different from ULS and some calculations are based on the reduced loads (SLS)

Author Response

Reviewer 2

Thank you for the review.

Two journal articles addressing the topic of the article are added to the reference list.

A section explaining the article layout is added.

Terminologies are defined.

Codes with the same load factors are simpler and therefore less prone to design errors, at least theoretically.

Timber VM = 0.2 is the optimum material or at least close to optimum for the primary material as seen in Figure 2, the material safety factors for VQ = 0.4 and VQ = 0.2 differ from each other quite little. If steel or concrete were selected, the corresponding deviation was bigger.

Gama_L is defined in line 100.

A short definition of Gumbel distribution is added.

In the article the characteristic load of the variable load is set changeable. This procedure is explained in sections 1.3, 1.4 and 3.2.

The difference between the dependent and the independent load combination is explained by the author in [17-20]. The material safety factors in the permanent loads are the same but the dependent load combination results about 20% higher material safety factors in the variable loads.

The dependent and the independent load combination is based on the same distributions. In author’s article “Test loading of structures with suspect resistance” https://www.mdpi.com/2076-3417/11/8/3424 the safety factors are given calculated for both methods.

References to equations are added.

A concise definition of the Monte Carlo simulation is added.

Wording in line 247 is made clearer.

Two equations are added which enable numerical checking of the results.

The selected coefficient of variations of the distributions commonly appear in the literature.

MATHCAD 15 program with limited graphical properties is used, the author tried to improve the quality of Figures.

The given material safety factors apply to the basic cases of the design i.e. the cases which are derived from the assumed distributions. Special cases are not addressed.

The structural analysis in the SLS and in the ULS is always the same in the statically determinate cases and in these cases one structural analysis is enough when the load factors are the same.

Round 2

Reviewer 1 Report

It seems that the Author addressed every comment/recommendation made by the Reviewer. The paper is now in a suitable form for publication.