message_ix_models.model.transport.config.Config

class message_ix_models.model.transport.config.Config(base_model_info: ~message_ix_models.util.scenarioinfo.ScenarioInfo = <factory>, constraint: ~typing.Dict = <factory>, cost: ~typing.Dict = <factory>, data_source: ~message_ix_models.model.transport.config.DataSourceConfig = <factory>, demand_modes: ~typing.List[str] = <factory>, dummy_demand: bool = False, dummy_supply: bool = False, efficiency: ~typing.Dict = <factory>, emission_relations: bool = True, factor: ~typing.Dict = <factory>, fast: bool = True, fixed_GDP: ~genno.core.attrseries.AttrSeries = <factory>, fixed_pdt: ~genno.core.attrseries.AttrSeries = <factory>, load_factor: ~typing.Dict = <factory>, lamda: float = -2.0, ldv_cost_catch_up_year: ~typing.Dict = <factory>, ldv_lifetime: ~genno.core.attrseries.AttrSeries = <factory>, minimum_activity: ~typing.Dict[~typing.Tuple[str, ~typing.Tuple[str, ...], str], float] = <factory>, mode_share: str = 'default', modules: ~typing.List[str] = <factory>, node_to_census_division: ~typing.Dict = <factory>, policy: bool = False, project: ~typing.Dict[str, ~enum.Enum] = <factory>, scaling: float = 1.0, share_weight_convergence: ~typing.Dict = <factory>, spec: ~message_ix_models.util.scenarioinfo.Spec = <factory>, speeds: ~genno.core.attrseries.AttrSeries = <factory>, ssp: ~message_ix_models.project.ssp.ssp_field | None = None, with_scenario: bool = False, with_solution: bool = False, work_hours: ~genno.core.attrseries.AttrSeries = <factory>, year_convergence: int = 2110, extra_modules: dataclasses.InitVar[typing.Union[str, typing.List[str]]] | None = None, futures_scenario: dataclasses.InitVar[str] | None = None, navigate_scenario: dataclasses.InitVar[str] | None = None)[source]

Bases: ConfigHelper

Configuration for MESSAGEix-Transport.

This dataclass stores and documents all configuration settings required and used by transport. It also handles (via from_context()) loading configuration and values from files like config.yaml, while respecting higher-level configuration, for instance model.Config.regions.

__init__(base_model_info: ~message_ix_models.util.scenarioinfo.ScenarioInfo = <factory>, constraint: ~typing.Dict = <factory>, cost: ~typing.Dict = <factory>, data_source: ~message_ix_models.model.transport.config.DataSourceConfig = <factory>, demand_modes: ~typing.List[str] = <factory>, dummy_demand: bool = False, dummy_supply: bool = False, efficiency: ~typing.Dict = <factory>, emission_relations: bool = True, factor: ~typing.Dict = <factory>, fast: bool = True, fixed_GDP: ~genno.core.attrseries.AttrSeries = <factory>, fixed_pdt: ~genno.core.attrseries.AttrSeries = <factory>, load_factor: ~typing.Dict = <factory>, lamda: float = -2.0, ldv_cost_catch_up_year: ~typing.Dict = <factory>, ldv_lifetime: ~genno.core.attrseries.AttrSeries = <factory>, minimum_activity: ~typing.Dict[~typing.Tuple[str, ~typing.Tuple[str, ...], str], float] = <factory>, mode_share: str = 'default', modules: ~typing.List[str] = <factory>, node_to_census_division: ~typing.Dict = <factory>, policy: bool = False, project: ~typing.Dict[str, ~enum.Enum] = <factory>, scaling: float = 1.0, share_weight_convergence: ~typing.Dict = <factory>, spec: ~message_ix_models.util.scenarioinfo.Spec = <factory>, speeds: ~genno.core.attrseries.AttrSeries = <factory>, ssp: ~message_ix_models.project.ssp.ssp_field | None = None, with_scenario: bool = False, with_solution: bool = False, work_hours: ~genno.core.attrseries.AttrSeries = <factory>, year_convergence: int = 2110, extra_modules: dataclasses.InitVar[typing.Union[str, typing.List[str]]] | None = None, futures_scenario: dataclasses.InitVar[str] | None = None, navigate_scenario: dataclasses.InitVar[str] | None = None) → None

Methods

`__init__`([base_model_info, constraint, ...])
`check`()	Check consistency of `project`.
`from_context`(context[, scenario, options])	Configure context for building MESSAGEix-Transport.
`from_dict`(data)	Construct an instance from data with name manipulation.
`read_file`(path[, fail])	Update configuration from file.
`replace`(**kwargs)	Like `dataclasses.replace()` with name manipulation.
`set_futures_scenario`(value)	Update `project` from a string indicating a Transport Futures scenario.
`set_navigate_scenario`(value)	Update `project` from a string representing a NAVIGATE scenario.
`update`(**kwargs)	Update attributes in-place.

Attributes

`dummy_demand`	Include dummy `demand` data for testing and debugging.
`dummy_supply`	Include dummy technologies supplying commodities required by transport, for testing and debugging.
`emission_relations`	Generate relation entries for emissions.
`extra_modules`	Extra entries for `modules`, supplied to the constructor.
`fast`	If `True` (the default), do not record/preserve parameter data when removing set elements from the base model.
`futures_scenario`	Identifier of a Transport Futures scenario, used to update `project` via `ScenarioFlags.parse_futures()`.
`lamda`	Logit share exponents or cost distribution parameters [0]
`mode_share`	Base year shares of activity by mode.
`navigate_scenario`	Identifiers of NAVIGATE T3.5 demand-side scenarios, used to update `project` via `ScenarioFlags.parse_navigate()`.
`policy`	indicates whether the base scenario used is a policy (carbon pricing) scenario, or not.
`scaling`	Scaling factors for production function [0]
`ssp`	Enum member indicating a Shared Socioeconomic Pathway, if any, to use for exogenous data.
`with_scenario`	`True` if a base model or MESSAGEix-Transport scenario (possibly with solution data) is available.
`with_solution`	`True` if solution data is available.
`year_convergence`	Year for share convergence.
`base_model_info`	Information about the base model.
`constraint`	Values for constraints.
`cost`	Scaling factors for costs.
`data_source`	Sources for input data.
`demand_modes`	Set of modes handled by demand projection.
`efficiency`	Various efficiency factors.
`factor`	Various other factors.
`fixed_GDP`	Fixed future point for total passenger activity.
`fixed_pdt`	Fixed future point for total passenger activity.
`load_factor`
`ldv_cost_catch_up_year`	Period in which LDV costs match those of a reference region.
`ldv_lifetime`	Lifetime of light duty vehicles.
`minimum_activity`	Tuples of (node, technology (transport mode), commodity) for which minimum activity should be enforced.
`modules`	List of modules containing model-building calculations.
`node_to_census_division`	Used by `get_USTIMES_MA3T()` to map MESSAGE regions to U.S.
`project`	Flags for distinct scenario features according to projects.
`share_weight_convergence`	Mapping from nodes to other nodes towards which share weights should converge.
`spec`	Specification for the structure of MESSAGEix-Transport, processed from contents of `set.yaml` and `technology.yaml`.
`speeds`	Speeds of transport modes.
`work_hours`	Work hours per year, used to compute the value of time.

base_model_info: ScenarioInfo: Information about the base model.

check()[source]: Check consistency of project.

constraint: Dict

Values for constraints.

“LDV growth_activity_lo”, “LDV growth_activity_up”: Allowable annual decrease or increase (respectively) in activity of each LDV technology. For example, a value of 0.01 means the activity may increase by 1% from one year to the next. For periods of length >1 year, MESSAGE compounds the value. Defaults are multiples of 0.0192 = (1.1 ^ 0.2) - 1.0; or ±10% each 5 years. See ldv.constraint_data().
“non-LDV growth_new_capacity_up”: Allowable annual increase in new capacity (roughly, sales) of each technology for transport modes other than LDV. See non_ldv.growth_new_capacity().
“* initial_*_up”: Base value for growth constraints. These values are arbitrary.

cost: Dict

Scaling factors for costs.

ldv nga

Scaling factor to reduce the cost of NGA vehicles.

Note

DLM: “applied to the original US-TIMES cost data. That original data simply seems too high - much higher than conventional gasoline vehicles in the base-year and in future, which is strange.

bus inv

Investment costs of bus technologies, relative to the cost of ICG_bus. Dictionary with 1 key per BUS technology.

Used in ikarus.py
This is from the IKARUS data in GEAM_TRP_Technologies.xlsx; sheet ‘updateTRPdata’, with the comment “Original data from Sei (PAO).”
This probably refers to some source that gave relative costs of different buses, in PAO, for this year; it is applied across all years.

data_source: DataSourceConfig: Sources for input data.

demand_modes: List[str]: Set of modes handled by demand projection. This list must correspond to groups specified in the corresponding technology.yaml file.

Todo

Read directly from technology.yaml

dummy_demand: bool = False: Include dummy demand data for testing and debugging.

dummy_supply: bool = False: Include dummy technologies supplying commodities required by transport, for testing and debugging.

efficiency: Dict: Various efficiency factors.

emission_relations: bool = True: Generate relation entries for emissions.

extra_modules: dataclasses.InitVar[Union[str, List[str]]] = None: Extra entries for modules, supplied to the constructor. May be either a space-delimited string ("module_a -module_b") or sequence of strings. Values prefixed with a hyphen ("-module_b") are removed from modules.

factor: Dict: Various other factors.

fast: bool = True: If True (the default), do not record/preserve parameter data when removing set elements from the base model.

fixed_GDP: AttrSeries: Fixed future point for total passenger activity.

fixed_pdt: AttrSeries

Fixed future point for total passenger activity.

AJ: Assuming mean speed of the high-speed transport is 330 km/h leads to 132495 passenger km / capita / year (Schafer & Victor 2000). Original comment (DLM): “Assume only half the speed (330 km/h) and not as steep a curve.”

classmethod from_context(context: Context, scenario: Scenario | None = None, options: Dict | None = None) → Config[source]

Configure context for building MESSAGEix-Transport.

If scenario is given, context.model.regions is updated to match. See identify_nodes().
context.transport is set to an instance of Config.

Configuration files and metadata are read and override the class defaults.

The files listed in METADATA are stored in the respective attributes, e.g. set corresponding to data/transport/set.yaml.

If a subdirectory of data/transport/ exists corresponding to context.model.regions then the files are loaded from that subdirectory, for instance data/transport/ISR/set.yaml is preferred to data/transport/set.yaml.

classmethod from_dict(data: Mapping): Construct an instance from data with name manipulation.

futures_scenario: dataclasses.InitVar[str] = None: Identifier of a Transport Futures scenario, used to update project via ScenarioFlags.parse_futures().

lamda: float = -2.0: Logit share exponents or cost distribution parameters [0]

ldv_cost_catch_up_year: Dict: Period in which LDV costs match those of a reference region. Dimensions: (node,).

ldv_lifetime: AttrSeries: Lifetime of light duty vehicles. Dimensions: (consumer type,). Units: [time].

minimum_activity: Dict[Tuple[str, Tuple[str, ...], str], float]: Tuples of (node, technology (transport mode), commodity) for which minimum activity should be enforced. See non_ldv.bound_activity_lo().

mode_share: str = 'default': Base year shares of activity by mode. This should be the stem of a CSV file in the directory data/transport/regions/mode-share/.

modules: List[str]: List of modules containing model-building calculations.

navigate_scenario: dataclasses.InitVar[str] = None: Identifiers of NAVIGATE T3.5 demand-side scenarios, used to update project via ScenarioFlags.parse_navigate().

node_to_census_division: Dict: Used by get_USTIMES_MA3T() to map MESSAGE regions to U.S. census divisions appearing in MA³T.

policy: bool = False

indicates whether the base scenario used is a policy (carbon pricing) scenario, or not. This currently does not affect any behaviour of transport except the selection of a base scenario via base_scenario_url().

Type:: Temporary setting for the SSP 2024 project

project: Dict[str, Enum]

Flags for distinct scenario features according to projects. In addition to providing values directly, this can be set by passing futures_scenario or navigate_scenario to the constructor, or by calling set_futures_scenario() or set_navigate_scenario() on an existing Config instance.

transport.build and transport.report code will respond to these settings in documented ways.

read_file(path: Path, fail='raise') → None

Update configuration from file.

Parameters:

path – to a .yaml file containing a top-level mapping.
fail (str) – if “raise” (the default), any names in path which do not match attributes of the dataclass raise a ValueError. Ottherwise, a message is logged.

replace(**kwargs): Like dataclasses.replace() with name manipulation.

scaling: float = 1.0: Scaling factors for production function [0]

set_futures_scenario(value: str | None) → None[source]

Update project from a string indicating a Transport Futures scenario.

See ScenarioFlags.parse_futures(). This method alters mode_share and fixed_demand according to the value (if any).

set_navigate_scenario(value: str | None) → None[source]

Update project from a string representing a NAVIGATE scenario.

See ScenarioFlags.parse_navigate().

share_weight_convergence: Dict: Mapping from nodes to other nodes towards which share weights should converge.

spec: Spec: Specification for the structure of MESSAGEix-Transport, processed from contents of set.yaml and technology.yaml.

speeds: AttrSeries: Speeds of transport modes. The labels on the ‘t’ dimension must match demand_modes. Source: Schäefer et al. (2010)

Note

Temporarily ignored for PR #551; data are read instead from speed.csv.

ssp: ssp_field = None: Enum member indicating a Shared Socioeconomic Pathway, if any, to use for exogenous data.

update(**kwargs)

Update attributes in-place.

Raises:: AttributeError – Any of the kwargs are not fields in the data class.

with_scenario: bool = False: True if a base model or MESSAGEix-Transport scenario (possibly with solution data) is available.

with_solution: bool = False: True if solution data is available.

work_hours: AttrSeries: Work hours per year, used to compute the value of time.

year_convergence: int = 2110: Year for share convergence.