2.3.2. Outputs

After the DL assessment is complete, the workflow collects the EDP.csv, DM.csv, and DV.csv files produced for each building asset and aggregates the results into single output files. When run locally, the workflow returns the output in CSV format, which is located in the results folder. When run remotely, the workflow returns the output in HDF format, which is located in the job archive. The HDF files can be converted into CSV format as a post-processing step.

2.3.2.1. EDP Output File

File which aggregates response simulation results from the EDP.csv files of all building assets.
properties
type	Name of EDP type. Typical earthquake EDPS are peak floor acceleration (PFA), peak interstory drift (PID), peak floor displacement (PFA), and peak roof displacement (PRD). If IMs are used as EDPs (with the ‘IMasEDP’ EDP application), then the IM name is returned in this field.
	type	string
loc	Location of where EDP is recorded on the model. If UserDefinedEDP is specified as the EDP application, then the location labelling is denoted in the EDP specification file. Otherwise, the numbering starts at 0, which denotes the ground level.
	type	int
dir	Direction of the EDP is recorded. The numbering used in the file is X=1, Y=2, Z=3.
	type	int
stat	Describes the statistic reported for each EDP. EDPs are assumed to follow a lognormal distribution described by a median parameter and beta (lognormal standard deviation) parameter. Note: If the ‘samples’ entry of the RegionalMapping application is set to 1, then only one event is assigned to each building asset and the response is reported as the ‘median’ in the EDP Output File.
	type	string

The header schema, followed by an example output, is provided:

Table 2.3.2.1 OutputEDPSchema.csv

type	PFA	PFA	PFA	PFA	PFD	PFD	PFD	PFD	PID	PID	PID	PID
loc	0	0	0	0	1	1	1	1	1	1	1	1
dir	1	1	2	2	1	1	2	2	1	1	2	2
stat	median	beta	median	beta	median	beta	median	beta	median	beta	median	beta
1	130.996	0.478197571	145.494	0.080967514	264.518	0.152547475	276.529	0.27654027	0.00182557	0.151875634	0.00190603	0.277128181

2.3.2.2. DM Output File

File which aggregates damage state estimates from the DM.csv files of all building assets.
properties
Collapse probability	Probability of collapse for each building asset, represented as a value between 0 and 1.
	type	float
DS likelihood	The probability of each damage state for every building asset, provided as a value between 0 and 1. Probabilities are reported for five damage states: ‘0’ indicates no damage; ‘1_1’ indicates DS1 (aesthetic damage), ‘2_1’ indicates DS2 (mild damage), ‘3_1’ indicates DS3 (moderate damage), and ‘4_1’ indicates DS4 (severe damage). The DS likelihoods are additionally reported by component group, where the following abbreviations are used: ‘S’ = structural components, ‘NS’ = all nonstructural components, ‘NSA’ = acceleration-sensitive nonstructural components, ‘NSD’ = drift-sensitive nonstructural components. The probabilities of the five damage states for each component type (across the row) should sum to 1 (total probability for the building asset).
	type	float

The header schema, followed by an example output, is provided:

Table 2.3.2.2 OutputDMSchema.csv

	Collapse	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood	DS likelihood
comp_type	probability	S	S	S	S	S	S	NS	NS	NS	NS	NS	NSA	NSA	NSA	NSA	NSA	NSD	NSD	NSD	NSD	NSD
DSG_DS		0	1_1	2_1	3_1	4_1	4_2	0	1_1	2_1	3_1	4_1	0	1_1	2_1	3_1	4_1	0	1_1	2_1	3_1	4_1
1	0	0.8	0.2						0.2	0.2	0.6			0.2	0.2	0.6		0.8	0.2

2.3.2.3. DV Output File

File which aggregates decision variable estimates from the DV.csv files of all building assets.
properties
Aggregate Repair Cost	Repair cost for the entire building asset, reported with the statistics: mean, standard deviation, 10th percentile, median (50th percentile), and 90th percentile. The currency unit for the repair cost is the same as that of the the replacementCost property provided in the building input file. If replacementCost is set equal to 1, then the repair costs in the DV Output File are reported as a percentage of the replacement cost.
	type	float
Repair Impractical	Probability of repair being impractical for the building asset, provided as a value between 0 and 1.
	type	float
Component Repair Cost	Mean repair cost for the building asset reported by component group (‘S’ = structural components, ‘NS’ = all nonstructural components, ‘NSA’ = acceleration-sensitive nonstructural components, ‘NSD’ = drift-sensitive nonstructural components) and damage state (‘aggregate’ = total for the component group, ‘1_1’ = DS1, ‘2_1’ = DS2, ‘3_1’ = DS3, ‘4_1’ = DS4).
	type	float
Aggregate Repair Time	Repair time for the entire building asset, reported with the statistics: mean, standard deviation, 10th percentile, median (50th percentile), and 90th percentile. Time is reported in units of ___.
	type	float
Aggregate Injuries	Number of injuries projected as a result of damage to the building asset, reported with the statistics: mean, standard deviation, 10th percentile, median (50th percentile), and 90th percentile. Injuries are provided at four severity levels.
	type	float

The header schema, followed by an example output, is provided:

Table 2.3.2.3 OutputDVSchema.csv

DV	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Impractical	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Cost	Repair Time	Repair Time	Repair Time	Repair Time	Repair Time	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries	Injuries
comp_type	aggregate	aggregate	aggregate	aggregate	aggregate	probability	S	S	S	S	S	S	NS	NS	NS	NS	NS	NSA	NSA	NSA	NSA	NSA	NSD	NSD	NSD	NSD	NSD						sev1	sev1	sev1	sev1	sev1	sev2	sev2	sev2	sev2	sev2	sev3	sev3	sev3	sev3	sev3	sev4	sev4	sev4	sev4	sev4
DSG_DS							aggregate	1_1	2_1	3_1	4_1	4_2	aggregate	1_1	2_1	3_1	4_1	aggregate	1_1	2_1	3_1	4_1	aggregate	1_1	2_1	3_1	4_1	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate	aggregate
stat	mean	std	10%	median	90%		mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	mean	std	10%	median	90%	mean	std	10%	median	90%	mean	std	10%	median	90%	mean	std	10%	median	90%	mean	std	10%	median	90%
1	0.056575	0.034084051	0.0138	0.08	0.086525	0	0.000725	0.003625					0.05585	0.005	0.027	0.082416667		0.0544	0.005	0.027	0.08		0.00145	0.00725				0.29	0.58	0	0	0.87	7.25E-05	0.000145	0	0	0.0002175	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0