cost_eco_model_linker.cost_calculations¶

Attributes¶

THIS_DIR

Functions¶

`cost_types`(cost, contingency, nsims)	Calculate key cost codes:
`initialize_cost_df`(years, nsims)	Initialize dataframe for storing sampled cost data.
`sample_cost_model`(nsims)	Sample cost model parameters for production, deployment, and LM models.
`lm_opex_distance_multiplier`(→ float)	Return the OPEX multiplier for LM interventions based on deployment distance.
`update_lm_factors`(lm_factors, iv_spec, pool_min, pool_max)	Update sampled LM factor dataframe with the year-specific larval release
`update_factors`(prod_factors, deploy_factors, iv_spec)	Update sampled cost model parameter dataframes with intervention specific parameters
`calc_production_requirement`(deploy_factors, prev_iv_spec)	Calculate the number of devices required in the previous intervention year
`calculate_costs`(stores, ID_key_fn, nsims, ...[, ...])	Sample costs for a set of interventions specified in ID_key, sampling nsims.

Module Contents¶

cost_eco_model_linker.cost_calculations.THIS_DIR¶

cost_eco_model_linker.cost_calculations.cost_types(cost, contingency, nsims)¶

Calculate key cost codes: - 1 : CAPEX, sum of production and deployment cost - 2 : Contingency CAPEX, % of CAPEX - 3 : OPEX, sum of production and deployment cost - 4 : Sustaining capital OPEX, set to zero for now (assumed to be included in OPEX through contract) - 5 : Contingency OPEX, % of OPEX - 6 : Vessel fuel, only relevant if volunteer vessels are used - set to zero for now - 7 : CAPEX-monitoring, set to zero (assumed no monitoring cost) - 8 : Contingency CAPEX-monitoring, % of CAPEX-monitoring - 9 : OPEX-monitoring, set to zero (assumed no monitoring cost) - 10 : Sustaining capital OPEX-monitoring, set to zero (assumed no monitoring cost) - 11 : Contingency OPEX-monitoring, % of OPEX-monitoring

Parameters:

cost (dataframe) – Dataframe containing ‘capex’ and ‘opex’
contingency (float) – Contingency proportion.
nsims (int) – Total number of simulations (from metrics sampling)

cost_eco_model_linker.cost_calculations.initialize_cost_df(years, nsims)¶

Initialize dataframe for storing sampled cost data.

Components: - 1 : CAPEX, sum of production and deployment cost - 2 : Contingency CAPEX, % of CAPEX - 3 : OPEX, sum of production and deployment cost - 4 : Sustaining capital OPEX, set to zero for now (assumed to be included in OPEX through contract) - 5 : Contingency OPEX, % of OPEX - 6 : Vessel fuel, only relevant if volunteer vessels are used - set to zero for now - 7 : CAPEX-monitoring, set to zero (assumed no monitoring cost) - 8 : Contingency CAPEX-monitoring, % of CAPEX-monitoring - 9 : OPEX-monitoring, set to zero (assumed no monitoring cost) - 10 : Sustaining capital OPEX-monitoring, set to zero (assumed no monitoring cost) - 11 : Contingency OPEX-monitoring, % of OPEX-monitoring

Parameters:

years (np.array) – All simulation years (non-intervention years will have zero costs)
nsims (int) – Total number of simulations (from metrics sampling)

Returns:

cost_df

Return type:

dataframe

cost_eco_model_linker.cost_calculations.sample_cost_model(nsims)¶

Sample cost model parameters for production, deployment, and LM models.

Parameters:

nsims (int or None) – Total number of simulations (from metrics sampling). When None or 1 Sobol’ sampling is skipped and a single row of best_point_value entries is returned instead.

Returns:

factor_specs_prod (DataFrame)
factors_df_prod (DataFrame)
factor_specs_dep (DataFrame)
factors_df_dep (DataFrame)
factor_specs_lm (DataFrame)
factors_df_lm (DataFrame)

cost_eco_model_linker.cost_calculations.lm_opex_distance_multiplier(distance_nm: float) → float¶

Return the OPEX multiplier for LM interventions based on deployment distance.

Derived from the regression 0.2495 * x^0.517, where x is distance in nautical miles. The relationship is calibrated so that 15 NM → ×1.0 (100%), 30 NM → ×1.5, 60 NM → ×2.0, 120 NM → ×3.0.

Parameters:: distance_nm (float) – Distance from port in nautical miles.
Returns:: Multiplier to apply to the raw LM OPEX value.
Return type:: float

cost_eco_model_linker.cost_calculations.update_lm_factors(lm_factors, iv_spec, pool_min, pool_max, sample_scale=False)¶

Update sampled LM factor dataframe with the year-specific larval release pool count derived from coral counts in the intervention specification.

Parameters:

lm_factors (DataFrame) – Factors dataframe for the LM cost model.
iv_spec (DataFrame) – Intervention specification for the current rep and year (one row per reefset); must contain number_of_1YO_corals.
pool_min (int) – Minimum valid pool count (range_lower from LM config).
pool_max (int) – Maximum valid pool count (range_upper from LM config).
sample_scale (bool, optional) – When True (cost exploration mode), the larval pool count is already sampled from the Sobol sequence and must not be overwritten. The RME template coral counts are ignored. Defaults to False.

Returns:

lm_factors

Return type:

DataFrame

cost_eco_model_linker.cost_calculations.update_factors(prod_factors, deploy_factors, iv_spec, sample_scale=False)¶

Update sampled cost model parameter dataframes with intervention specific parameters for a single ecological repeat.

Parameters:

prod_factors (dataframe) – Factors dataframe for the production cost model.
deploy_factors (dataframe) – Factors dataframe for the deployment cost model.
iv_spec (dataframe) – Intervention specification dataframe for the current rep and year, containing one row per reefset.
sample_scale (bool, optional) – When True (cost exploration mode), num_1yoec is already sampled from the Sobol sequence and must not be overwritten with the fixed coral count from the RME template. coral_yield_1YOEC is still synced between models and distance_from_port is still set from iv_spec. Defaults to False.

cost_eco_model_linker.cost_calculations.calc_production_requirement(deploy_factors, prev_iv_spec)¶

Calculate the number of devices required in the previous intervention year for a single ecological repeat. Used to determine whether additional production capacity is needed in the current year.

Parameters:

deploy_factors (dataframe) – Factors dataframe for the deployment cost model (used for yield values).
prev_iv_spec (dataframe) – Intervention specification dataframe for the previous intervention year and rep, containing one row per reefset.

Returns:

prev_devices – Number of devices required in the previous intervention year, per simulation.

Return type:

np.ndarray

cost_eco_model_linker.cost_calculations.calculate_costs(stores: cost_eco_model_linker.setup_results.OutputStores, ID_key_fn: str, nsims: int, deploy_model_filepath: str, prod_model_filepath: str, lm_model_filepath: str, cont_p: float = 0.25, p_iter_id: int = 0, active_models: set = None, sample_scale: bool = False, session_reset_interval: int = _RESET_INTERVAL)¶

Sample costs for a set of interventions specified in ID_key, sampling nsims.

Parameters:

stores (OutputStores) – Data class holding output directory locations
ID_key_fn (str) – Target filename for output.
nsims (int) – Total number of draws to sample cost models, should match ecological metrics sampling.
deploy_model_filepath (string) – Path to deployment cost model.
prod_model_filepath (string) – Path to production cost model.
lm_model_filepath (string) – Path to LM cost model.
cont_p (float) – Contingency cost proportion.
p_iter_id (int) – ID used for parallel sampling to keep track of batches for ordered recombination.
active_models (set, optional) – Set of intervention types to compute costs for. Valid values are "outplant" and "lm". Defaults to both when None. Pass {"outplant"} for CA-only or {"lm"} for LM-only scenarios.
sample_scale (bool, optional) – When True (cost exploration mode), intervention scale (num_1yoec for outplant, larval pool count for LM) is drawn from the Sobol samples rather than derived from the RME template coral counts. Defaults to False.
session_reset_interval (int, optional) – Recycle the process-local Excel session every N calls to bound memory growth. Defaults to _RESET_INTERVAL (50).