Inquiry on Code Level Mapping, Fragility Function Usage, and Intensity Measure Sampling in R2D

[MohamedKhoudari]

Dear Jinyan,

First of all, thank you for your continuous support.

I have a few inquiries I would appreciate your guidance on:

1. I noticed that the Low Code (LC) design level appears to be the lowest code level assigned in R2D to defined buildings. Even when a building is assigned a construction year from the late 1800s or early 1900s, R2D still classifies it as LC, rather than Pre-Code (PC).
   I am following the R2D E-14 Basic HAZUS transportation example template in my analysis. I reviewed the fragility function files, namely “damage_DB_Hazus_EQ_bldg” and “damage_DB_Hazus_EQ_trnsp”, which I assume R2D uses during analysis. These files include fragility functions for Pre-Code (PC) buildings. Therefore, I would like to know:
   • Is there a specific approach, such as assigning a particular year of construction, to ensure R2D applies the PC fragility functions?
   • Could you also share the mapping between the year of construction and the assigned design code levels as used in R2D?

2. I provided an Intensity Measures (IMs) file similar to that in E-14, which includes PGA, Sa(0.3s), Sa(1.0s), PGD_h, and PGD_v. Based on the results, it appears R2D applies PGA-dependent fragility functions (e.g., LF.RES3-LF.C1.M.LC) for evaluating building damage. Could you please clarify under what conditions, if any, Sa(0.3s) and Sa(1.0s) are used?
   As for roads, R2D employs PGD-based fragility functions (e.g., HRD-HRD.GF.2). I would like to confirm which PGD component (horizontal or vertical) is used in the analysis.

3. Lastly, when R2D selects a set of intensity measures from neighboring sites for asset mapping, does it select all IMs from the same row, maintaining the correlation across IMs per site and the site-to-site correlation, or are IMs for each parameter (PGA, Sa, PGD) independently sampled from their respective columns? This point is crucial for me, as the IMs I provided are correlated per row.

Thank you again for your valuable support.

Best regards,
Mohamed Khoudari

[jinyan1214]

Hi Mohamed,

Thanks for using R2D, and it is my pleasure to support. Please see comments below:

1. In R2D, the Python package Pelicun is used to map between each asset and its fragility and perform sampling according to the fragilities. If you are using the latest R2D and associated Pelicun, the mapping rule for buildings is here. In the latest version, design_level is not inferred from construct year. Rather, the user provided a "DesignLevel" attribute for each building in the asset inventory you feed R2D, with possible DesignLevel values suggested on line 114 of the pelicun_config.py. In older versions of R2D and pelicun, the inference of DesignLevel and many others was performed in pelicun_config.py. In current and future versions, these inference rules are moved to BRAILS++. In summary, to use your desired design level, please update to the latest R2D and Pelicun, add a "DesignLevel" attribute to your building inventory, and Pelicun should pick up the appropriate design level.
2. The Sa(0.3s) and Sa(1.0s) are used if the "Capacity Spectrum Method" is selected for building analysis in the "ANA" panel. This analysis performs a capacity spectrum analysis according to FEMA's HAZUS technical manual. The capacity spectrum method produces a drift ratio and a peak floor acceleration for a building, and the corresponding fragility curves listed here are used. Besides, the fragility curves for highway bridges require Sa(0.3s) and Sa(1.0s). As for roads, R2D uses the bigger one between the PGD_h and PGD_v in each realization for fragility analysis.
3. Yes, R2D selects all IMs from the same row to maintain correlations.

Best regards,

Jinyan