Review Reports

Round 1

Reviewer 1 Report

Brief Summary

The paper presents a study on natural hazard influence on maintenance and repair costs of accommodation facilities. The data used in the study comes from a hotel chain insurer and contains all the claims during a period of 10 years. A regression model is used to find a correlation between natural phenomenon and insurance-claim payouts. The independent variables chosen for the model are: wind speed, peak ground acceleration (PGA), precipitation, distance to water systems and altitude difference to water systems, the dependent variable is the hotel chain gross loss. The paper concludes giving a relatively low adjusted R2 which explains a 34.2% of the variation of the dependent variable function of the independent variables. Most of this correlation comes from the influence of precipitation to the gross loss.

 

Broad comments

The abstract mentions the climate change and increase of high-rise and high-performance buildings as a source of increased maintenance costs, nevertheless these two factors are not mentioned anymore during the paper, why is the approach developed in this paper more suitable to this new paradigm?

Ref [16] uses artificial neural networks (ANNs) for cost estimation; do you think that this approach could improve the choice of independent variables in the present study?

In section 3: research methods, a deeper description of the statistics used is needed. The paper is using a multiple linear regression to correlate the chosen independent variables with the dependent variable, to perform such statistical analysis some requirements are needed: linear relationship between dependent variable and independent variables, homoscedasticity, independent variables must not show multicollinearity, outliers need to be filtered, errors are approximately normally distributed. Does the data fulfill all the requirements?

The paper contains no plots at all, a scatter plot showing the dependent variable against each of the independent ones would be very helpful to visualize the distribution of the data, presence of outliers, check for linearity, etc.

In line 160: ‘According to Harvard’s Joint Center for Housing Studies, among natural disasters, earthquakes and hurricanes are most costly for residential buildings’, nevertheless in the results presented wind speed and PGA have no correlation with damage costs or even a slight negative correlation. It is true that Ref. [28] refers to residential buildings, not hotels and an important factor in hurricane damage is precipitation but still, having no correlation between wind speed, PGA and damage cost is rather puzzling. How do you explain it?

The two independent variables: distance and altitude to water systems show no correlation with damage costs. I find this not surprising and I think that other variables are much more significant to flooding risks; did you consider using Rational Method variables instead? Like basin area, vegetation, slope, rain intensity factors, etc. Would it be challenging to define those for the present study?

 

Line by line comments

Line 44-47: Yet, despite … their management tends to be outdated, passive and unsystematic. This statement needs a reference.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper addresses a very interesting subject. Based on eleven years of data on an international hotel chain’s insurance-claim payout, the study aimed to identify the correlations among maintenance costs, damage, and the incidence and intensity of earthquakes, high winds, and flooding.

The text is well written and well structured. The motivation and the importance of the study are clearly presented. On the other hand, the results and discussion are quite disappointing. Although a lot of work is still needed to confirm the main findings, in the reviewer’s opinion the paper can be accepted for publication if some aspects are addressed, being:

-The title and some parts of the paper suggest the consideration of sustainability as a whole, but social and environmental costs were not taken into account. This must be emphasized in the introduction;

-The average distance of water systems is about 55 km! How is this aspect so relevant?

- More information about data collection would be interesting to the readers, such as the percentual of each kind of damage concerning total costs;

-Line 255: “Nevertheless, future research should incorporate more sources and types of damage data to confirm the present findings”. Please, exemplify.

-Line 266: “facility-management companies could use the results of the present research as a basis for hazard mapping or hazard-prediction modeling aimed at managing their properties more effectively by reducing unexpected costs”. How can they do this?

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors correctly address the remarks of the first review report. The reviewer thanks the authors for the attention devoted to the revision.

1. The introduction now links more smoothly the climate change, new construction paradigms and damage costs treated in the paper. New bibliography helps in this aspect.
2. Possible future use of artificial neural networks (ANN) for investigating variables relationships is now introduced in the literature review.
3. This was the main aspect needing major revision, with the new text, figures and tables the paper is much more sound and reproducible.
4. Related to previous item.
5. Now it is clear and the citation is no more in contradiction with the results.
6. This is clearer in the revised version and helps to reply to the other reviewer’s remark concerning mean distance to water systems.
7. Ok.

     

Reviewer 2 Report

The authors have satisfactorily responded to my questions and made the necessary changes to the manuscript.