Skip to contents

Computes the mode shares for the different distance categories and then the maximum mode share for specified mode types and specified distance categories across specified (stored) cities. Used for max mode share scenario generation.

Usage

get_scenario_settings(
  cities = c("accra", "bangalore", "belo_horizonte", "bogota", "buenos_aires",
    "cape_town", "delhi", "mexico_city", "santiago", "sao_paulo", "vizag"),
  modes = c("pedestrian", "cycle", "car", "motorcycle", "bus"),
  distances = c("0-2 km", "2-6 km", "6+ km"),
  speeds = list(bus = 8.1, bus_driver = 8.1, minibus = 8.1, minibus_driver = 8.1, car =
    13.8, car_driver = 13.8, taxi = 13.8, pedestrian = 2.5, walk_to_pt = 2.5, cycle =
    7.2, motorcycle = 15.2, truck = 8.1, van = 13.8, subway = 18.1, rail = 21.9,
    auto_rickshaw = 4, shared_auto = 13.8, shared_taxi = 13.8, cycle_rickshaw = 4, other
    = 9.1)
)

Arguments

cities

which cities to use

modes

which modes to use

distances

which distance categories to use

speeds

named list of mode speeds (to be applied to all cities)

Value

data frame of maximum proportions by mode and distance category

Details

The function performs the following steps:

  • define the minimum distances in each distance category

  • loop through the pre-defined cities:

    • read in trip data and get data into correct format

    • assign distance categories

    • for each distance category and pre-defined mode find the proportional modal share (for each distance category, the proportion of the modes adds up to 100

    • if rail trips exist, they are added to the proportion of the bus trips to get one value for public transport

  • find the maximum mode shares for each mode and distance category

  • for each city print one table showing the mode shares for each distance category