Inconsistent building types in samples and postprocessing

[JieXiong9119]

Description

Feb/2024 (EUSS 2024c1)

The office building subcategory ("LargeOffice","MediumOffice","SmallOffice") for in.comstock_building_type should follow the logic and definition here:

ComStock/postprocessing/comstockpostproc/cbecs.py

Lines 309 to 327 in a541f15

	# Assign size category to offices
	if cstock_bldg_type == 'Office':
	if sqft < 25_000:
	if nfloor <=3:
	cstock_bldg_type = 'SmallOffice'
	else:
	cstock_bldg_type = 'MediumOffice'
	elif sqft >= 25_000 and sqft < 150_000:
	if nfloor <= 5:
	cstock_bldg_type = 'MediumOffice'
	else:
	cstock_bldg_type = 'LargeOffice'
	elif sqft >= 150_000:
	cstock_bldg_type = 'LargeOffice'
	else:
	err_msg = f"Should never get here, check logic for {row}"
	logger.error(err_msg)
	raise Exception(err_msg)
	assert cstock_bldg_type != 'Office' # Offices must be assigned a size

However when performing applicability check for "LargeOffice" for demand flexibility measures, it resulted in inconsistent office building types compared to metadata/samples. Here is a summary for a check performed on the comstock_baseline_metadata_and_annual_results.csv released in cycle 2:

---

total office building # in meta file:
74179

large office # in meta file:
4200
large office # by check:
3216
matched large office #:
2721
large office by check but not large office in meta file:
495
large office in meta file but not large office by check:
1479

medium office # in meta file:
7160
medium office # check:
7176
matched medium office #:
5791
medium office by check but not medium office in meta file:
1385
medium office in meta file but not medium office by check:
1369

small office # in meta file:
62819
small office # check:
63787
matched small office #:
62707
small office by check but not small office in meta file:
1080
small office in meta file but not small office by check:
112

Jan/2025 - Outpatient building showing Large Office in models

DF measures are now applicable to offices, schools and warehouse. However the run is showing 899 outpatient samples in this EUSS (SDR) cycle's (2024c2) results:

Quick check on results shows that all these outpatient models are located in California. Suspect the error has to do with DEER building type mapping when generating the models.
I downloaded one of the outpatient models showing in DF results and indeed found conflicting building type definition within the osm model:

Suggestion

Please confirm the generation of in.comstock_building_type field is correct or the logic and definition in postprocessing match the source. Please check the model articulation step is correct when dealing with DEER building type