blocksnet.method.annealing_optimizer

Module providing class for simulated annealing optimization of blocks services constrained by land uses.

Classes

`AnnealingOptimizer`(*, city_model[, verbose, ...])	Services optimizer using simulated annealing to maximize scenario provision objective function by building certain services.
`Indicator`(block, land_use, fsi, gsi)	Represents indicators for a block based on land use and spatial indices.
`Variable`(block, service_type, brick[, value])	Represents a variable for optimization in a given block with a specific service type and brick.

class blocksnet.method.annealing_optimizer.Variable(block: Block, service_type: ServiceType, brick: ServiceBrick, value: int = 0)[source]

Bases: object

Represents a variable for optimization in a given block with a specific service type and brick.

Variables:

block (Block) – The block where the service is located.
service_type (ServiceType) – The type of service provided.
brick (ServiceBrick) – The brick representing the service’s characteristics.
value (int, optional) – The quantity of the service brick, by default 0.

property capacity

Calculates the total capacity of the service brick in the block.

Returns:: The total capacity of the service brick.
Return type:: int

property area

Calculates the total area occupied by the service brick.

Returns:: The total area occupied by the service brick.
Return type:: int

to_dict()[source]

Converts the Variable instance to a dictionary representation.

Returns:: A dictionary containing the block ID, service type, brick capacity, brick area, whether the brick is integrated, and the value of the brick.
Return type:: dict

class blocksnet.method.annealing_optimizer.Indicator(block: Block, land_use: LandUse, fsi: float, gsi: float)[source]

Bases: object

Represents indicators for a block based on land use and spatial indices.

Variables:

block (Block) – The block being evaluated.
land_use (LandUse) – The type of land use for the block.
fsi (float) – The Floor Space Index (FSI) of the block.
gsi (float) – The Ground Space Index (GSI) of the block.

property site_area

Gets the site area of the block.

Returns:: The site area of the block.
Return type:: float

property footprint_area

Calculates the footprint area of the block based on the GSI.

Returns:: The footprint area of the block.
Return type:: float

property build_floor_area

Calculates the buildable floor area of the block based on the FSI.

Returns:: The buildable floor area of the block.
Return type:: float

property integrated_area

Determines the integrated area of the block based on land use.

Returns:: The integrated area of the block.
Return type:: float

property non_integrated_area

Calculates the non-integrated area of the block.

Returns:: The non-integrated area of the block.
Return type:: float

property living_area

Determines the living area of the block based on land use.

Returns:: The living area of the block. Returns 0 if land use is not residential.
Return type:: float

property population

Estimates the population of the block based on living area and demand.

Returns:: The estimated population of the block.
Return type:: int

to_dict() → dict[source]

Converts the Indicator instance to a dictionary representation.

Returns:: A dictionary containing block ID, site area, land use, footprint area, build floor area, integrated area, non-integrated area, living area, and population.
Return type:: dict

class blocksnet.method.annealing_optimizer.AnnealingOptimizer(*, city_model: City, verbose: bool = True, on_iteration: Callable[[int, list[blocksnet.method.annealing_optimizer.Variable], dict[int, blocksnet.method.annealing_optimizer.Indicator], float], None] | None = None)[source]

Bases: BaseMethod

Services optimizer using simulated annealing to maximize scenario provision objective function by building certain services.

Variables:: on_iteration (Callable[[int, list[Variable], dict[int, Indicator], float], None] | None) – A callback function to be called after each iteration with the current iteration number, the current list of variables, indicators, and the best value found so far.

on_iteration: Callable[[int, list[blocksnet.method.annealing_optimizer.Variable], dict[int, blocksnet.method.annealing_optimizer.Indicator], float], None] | None

_check_constraints(X, indicators) → bool[source]

Checks if the current solution satisfies all constraints.

Parameters:

X (list[Variable]) – List of variables representing the current solution.
indicators (dict[int, Indicator]) – Dictionary mapping block IDs to their corresponding indicators.

Returns:

True if all constraints are satisfied, False otherwise.

Return type:

bool

static _perturb(X: list[blocksnet.method.annealing_optimizer.Variable]) → tuple[list[blocksnet.method.annealing_optimizer.Variable], blocksnet.models.service_type.ServiceType][source]

Perturbs the current solution by modifying a randomly chosen variable.

Parameters:: X (list[Variable]) – List of variables representing the current solution.
Returns:: A tuple containing the perturbed list of variables and the service type of the perturbed variable.
Return type:: tuple[list[Variable], ServiceType]

_generate_initial_X(blocks_lu: dict[int, blocksnet.models.land_use.LandUse], service_types: dict[str, float]) → list[blocksnet.method.annealing_optimizer.Variable][source]

Generates an initial list of variables for the optimization.

Parameters:

blocks_lu (dict[int, LandUse]) – Dictionary mapping block IDs to their corresponding land uses.
service_types (dict[str, float]) – Dictionary mapping service type names to their weights.

Returns:

A list of variables representing the initial solution.

Return type:

list[Variable]

_generate_indicators(blocks, fsis, gsis) → dict[int, blocksnet.method.annealing_optimizer.Indicator][source]

Generates indicators for each block based on the provided data.

Parameters:

blocks (dict[int, LandUse]) – Dictionary mapping block IDs to their corresponding land uses.
fsis (dict[int, float]) – Dictionary mapping block IDs to their Floor Space Index (FSI) values.
gsis (dict[int, float]) – Dictionary mapping block IDs to their Ground Space Index (GSI) values.

Returns:

A dictionary mapping block IDs to their corresponding indicators.

Return type:

dict[int, Indicator]

to_gdf(X: list[blocksnet.method.annealing_optimizer.Variable], indicators: dict[int, blocksnet.method.annealing_optimizer.Indicator]) → GeoDataFrame[source]

Converts the list of variables and indicators into a GeoDataFrame.

Parameters:

X (list[Variable]) – List of variables representing the current solution.
indicators (dict[int, Indicator]) – Dictionary mapping block IDs to their corresponding indicators.

Returns:

A GeoDataFrame containing optimization result data.

Return type:

gpd.GeoDataFrame

to_bricks_df(X: list[blocksnet.method.annealing_optimizer.Variable]) → DataFrame[source]

Converts the list of variables into a DataFrame representing the bricks.

Parameters:: X (list[Variable]) – List of variables representing the current solution.
Returns:: A DataFrame containing data about the bricks in the solution.
Return type:: pd.DataFrame

_get_clear_df(blocks: list[int]) → DataFrame[source]

Constructs a DataFrame for provision assessment so the blocks being changed are treated as cleared.

Parameters:: blocks (list[int]) – List of changing blocks ids.
Returns:: A DataFrame containing information related to blocks being changed for provision assessment.
Return type:: pd.DataFrame

to_df(X: list[blocksnet.method.annealing_optimizer.Variable], indicators: dict[int, blocksnet.method.annealing_optimizer.Indicator]) → DataFrame[source]

Converts the list of variables and indicators into a DataFrame.

Parameters:

X (list[Variable]) – List of variables representing the current solution.
indicators (dict[int, Indicator]) – Dictionary mapping block IDs to their corresponding indicators.

Returns:

A DataFrame containing data about the blocks and their service capacities.

Return type:

pd.DataFrame

calculate(blocks_lu: dict[int, blocksnet.models.land_use.LandUse], blocks_fsi: dict[int, float], blocks_gsi: dict[int, float], service_types: dict[str, float], t_max: float = 100, t_min: float = 0.001, rate: float = 0.95, max_iter: int = 1000) → tuple[source]

Executes the services optimization process based on simulated annealing for provided service types weights and blocks landuses.

Parameters:

blocks_lu (dict[int, LandUse]) – Dictionary mapping block IDs to their corresponding land uses.
blocks_fsi (dict[int, float]) – Dictionary mapping block IDs to their Floor Space Index (FSI) values.
blocks_gsi (dict[int, float]) – Dictionary mapping block IDs to their Ground Space Index (GSI) values.
service_types (dict[str, float]) – Dictionary mapping service type names to their weights.
t_max (float, optional) – The maximum temperature for the annealing process, by default 100.
t_min (float, optional) – The minimum temperature for the annealing process, by default 1e-3.
rate (float, optional) – The cooling rate of the temperature, by default 0.95.
max_iter (int, optional) – The maximum number of iterations, by default 1000.

Returns:

A tuple containing the best solution found, the indicators, the best value, and the provision values for each service type.

Return type:

tuple

model_computed_fields: ClassVar[Dict[str, ComputedFieldInfo]] = {}: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.