Scenario creation • MyFirstFourStepModel

The code in the Classroom Implementation vignette runs a complete, simplified, four-step model for baseline conditions, as well as pre-specified land use and transportation network scenarios. This vignette describes how to create those land use and transportation network scenarios. This may be used by an instructor after estimating the model for a new region, or by advanced students. When I use the model in more advanced classes, I add an exercise that involves creating a land use scenario.

Land use scenarios

There are two ways to create new land use scenarios. The simpler method is to use the add_households function. This function modifies a scenario by adding households with particular characteristics to a particular area. The code below adds 20,000 households to Census tract 37037020801, which is the location of a large new residential development. The households to add are specified in a tabular format. In this case, are all four-person, two-worker households; half have three cars and income of $150,000/year, and half have two cars and income of $75,000/year. Household size can take on values from 1 to 4; workers and vehicles can take on values from 0 to 3, and income can take on values 0, 35000, 75000, and 100000, each representing the low end of an income bin. Every model includes a default “baseline” scenario which represents current conditions from the American Community Survey; other scenarios can be built by extending or modifying this baseline.

model$scenarios$chatham_park = model$scenarios$baseline |>
  add_households(
    "37037020801",
    tibble::tribble(
      ~hhsize , ~workers , ~vehicles , ~income , ~n    ,
            4 ,        2 ,         3 ,  150000 , 10000 ,
            4 ,        2 ,         2 ,   75000 , 10000
    )
  )

If you want to add households to more than one tract, simply use add_households more than once to build up the scenario one piece at a time.

If more control is needed, scenarios can be created in a tabular format by hand or using external tools. This format is shown in @tbl-demographic-scenario, specifying the number of households in different household size, income, vehicle ownership, and number of worker categories. It is up to the user to ensure that number of households in a tract is consistent across all marginal categories (i.e. if the four hhsize categories total 1000 households, the four income categories must also total 1000 households).

Specification of a demographic scenario {#tbl-demographic-scenario}
`geoid`	`marginal`	`value`	`count`
37183053411	`hhsize`	1	514
37183053411	`hhsize`	2	711
37183053411	`hhsize`	3	940
37183053411	`hhsize`	4	1907
37183053411	`income`	0	358
37183053411	`income`	35000	595
37183053411	`income`	75000	183
37183053411	`income`	100000	2936
37183053411	`vehicles`	0	110
37183053411	`vehicles`	1	921
37183053411	`vehicles`	2	2089
37183053411	`vehicles`	3	952
37183053411	`workers`	0	288
37183053411	`workers`	1	1784
37183053411	`workers`	2	1711
37183053411	`workers`	3	289

To create such a scenario, first export the baseline scenario into Excel format:

save_landuse_scenario(model$scenarios$baseline, "scenario.xlsx")

You can then modify the scenario using external tools. In addition to the tab specifying household counts, there is also a tab specifying employment in each tract, which is used to generate attractions. The column names match those in the LEHD LODES dataset. Once you are done modifying the scenario, you can load it back into R by running:

model$scenarios$projected = load_landuse_scenario("scenario.xlsx")

You can replace projected with any name you want (as long as it is a valid R identifier, e.g. no spaces).

Any scenarios created before the model is saved (see Estimation) will be available for use when the model is loaded.

Network scenarios

Changes to the network are just as important as changes to land use. Currently, My First Four-Step Model only supports changes to existing links.

Two link attributes can be changed: the lane count and the roadway type. The roadway type uses OpenStreetMap highway tag taxonomy; notable values are ‘motorway’, ‘trunk’, and ‘primary’. Changing the attributes of a link requires determining its OpenStreetMap ‘way ID’, which can be done either by looking at the data on openstreetmap.org, or exporting the network to GIS format and investigating in GIS:

network_to_gis(model$networks$baseline, "baseline.gpkg")

Link attributes can then be modified using the modify_links function. For example, the code below widens US 15-501 north of Pittsboro, NC, from two to three lanes in each direction, and upgrades it to a motorway.

model$networks$widen_15_501 = model$networks$baseline |>
  modify_ways(
    # US 15-501 between Pittsboro and Chapel Hill
    c(
      "16468788",
      "1435835324",
      "16471803",
      "1428927957",
      "16476716",
      "1435843899",
      "1435833214",
      "29335841",
      "709833467",
      "29335943",
      "654023608",
      "654023604",
      "29336020",
      "1428519228",
      "29336043",
      "29336065",
      "1119560022",
      "690900371",
      "133051279",
      "29336072",
      "690900390",
      "29336325",
      "29336326",
      "29336327",
      "29336328",
      "29336338",
      "29336335",
      "29336336",
      "133051274",
      "133051276",
      "133051275",
      "133051277",
      "1435849758",
      "133051278",
      "1119560050",
      "690900383",
      "138138830",
      "1064169646",
      "285898976",
      "625793296",
      "285898977",
      "712336832",
      "285898984",
      "654023612",
      "285898992",
      "1424926581",
      "1423515894",
      "1426579770",
      "1423515897",
      "1423515898",
      "1423515900",
      "398223958",
      "712336806",
      "398223959",
      "712336808",
      "1424926579",
      "1425787647",
      "1424926580",
      "1426816070",
      "1424926582",
      "518951244",
      "709833464",
      "713044971",
      "1425787648",
      "1425792246",
      "1426357461",
      "574612704",
      "614242454",
      "1426357462",
      "1426357465",
      "614242450",
      "614242453",
      "1426579771",
      "1426816072",
      "1428519229",
      "1428927958",
      "690900386",
      "690900353",
      "690900357",
      "1428247555",
      "1428247554",
      "694843964",
      "1425792245",
      "694843965",
      "1423515895",
      "1426357464",
      "709833465",
      "709833466",
      "998595933",
      "712336807",
      "712336820",
      "712336809",
      "712336821",
      "712336826",
      "712336827",
      "998595932",
      "822063218",
      "1435833215",
      "1435833216",
      "1435835323",
      "1435835325",
      "1435851262",
      "1435851263",
      "1028209511",
      "1435849757",
      "1064169630",
      "1064169631",
      "1064169647",
      "1265931335"
    ),
    lanes_per_direction = 3,
    highway_type = "motorway"
  )