Introduce Vacant Units

[afontani]

Pull Request Description

Companion Pull Request: resstock-estimation #36

The purpose of the pull request is to introduce vacant units into the residential stock. This work was motivated by the following maps that shows the fraction of vacant units by building type and PUMA.

The large fraction of vacant units, especially multi-family units in urban areas may partially be causing our over prediction of the electric load from multi-family dwelling units according to timeseries Load Research Data (LRD) or Advanced Metering Infrastructure (AMI) data. Utilities typically do not tract vacant units, just metered units or customers. The introduction of vacant units should help bring down the average or aggregate load per dwelling unit.

Data identified

Data was pulled from PUMS 5-yr 2016. PUMS distinguishes the number of vacant dwelling units and what type of unit is vacant. The fields from PUMS are as follows:

    0: 'N/A', # Occupied Unit
    1: 'For rent or sale',
    2: 'For sale only',
    3: 'Rented or sold but not (yet) occupied',
    4: 'For seasonal, recreational or other occasional use',
    5: 'For occasional use',
    6: 'For seasonal use',
    7: 'For migrant farm workers',
    8: 'For seasonal use or migratory',
    9: 'Other vacant'

Vacancy Status.tsv

These delineations are put into 2 options: Occupied and Vacant. Where sample counts are less than 10, the state average distribution is inserted.

Assumptions

Seasonal or occasional use units

Currently in options_lookup.tsv Seasonal or occasional use units are treated as vacant. The option is still split out for future improvement if/when vacations are includes into the stochastic occupancy model or when data is identified to determine the when the seasonal or occasional use units are occupied.

Heating and Cooling Setpoints and Setpoint Schedules

It is currently assumed that the heating and cooling setpoints and setpoint schedules are the same as occupied units.

YAML file update

With the introduction of vacant units, the total units represented should be updated to the total dwelling units in the American Community Servey (ACS). The total dwelling units in ACS 5-yr 2016 is 133,172,057.

Checklist

Not all may apply:

• Unit tests have been added or updated
• All rake tasks have been run, and pass
• Documentation has been modified appropriately
• Any new options are added to project_testing
• project_testing runs without any failures
• No unexpected regression test changes
• All tests are passing (green) on circleci
• The changelog has been updated appropriately
• This branch is up-to-date with master

For more information on how to perform these checklist items, see the documentation's Advanced Tutorial.