aeromaps.models.air_transport.air_traffic.price_elasticity

price_elasticity

Module for computing air traffic (RPK) with price elasticity effects.

RPKWithElasticity

RPKWithElasticity(name='rpk_with_elasticity', *args, **kwargs)

Bases: AeroMAPSModel

Class to compute Revenue Passenger Kilometers (RPK) with price elasticity and COVID-19 impact, considering exogenous growth rates by segment.

Parameters:

Name	Type	Description	Default
name		Name of the model instance ('rpk_with_elasticity' by default).	'rpk_with_elasticity'

Source code in aeromaps/models/air_transport/air_traffic/price_elasticity.py

def __init__(self, name="rpk_with_elasticity", *args, **kwargs):
    super().__init__(name=name, *args, **kwargs)

compute

compute(rpk_init, short_range_rpk_share_2019, medium_range_rpk_share_2019, long_range_rpk_share_2019, covid_start_year, covid_rpk_drop_start_year, covid_end_year_passenger, covid_end_year_reference_rpk_ratio, cagr_passenger_short_range_reference_periods, cagr_passenger_short_range_reference_periods_values, cagr_passenger_medium_range_reference_periods, cagr_passenger_medium_range_reference_periods_values, cagr_passenger_long_range_reference_periods, cagr_passenger_long_range_reference_periods_values, rpk_short_range_measures_impact, rpk_medium_range_measures_impact, rpk_long_range_measures_impact, airfare_per_rpk, price_elasticity)

Execute the computation of prospective air traffic.

Parameters:

Name	Type	Description	Default
rpk_init	Series	Historical number of Revenue Passenger Kilometer (RPK) over 2000-2019 [RPK].	required
short_range_rpk_share_2019	float	Share of RPK from short-range market in 2019 [%].	required
medium_range_rpk_share_2019	float	Share of RPK from medium-range market in 2019 [%].	required
long_range_rpk_share_2019	float	Share of RPK from long-range market in 2019 [%].	required
covid_start_year	Number	Covid-19 start year [yr].	required
covid_rpk_drop_start_year	float	Drop in RPK due to Covid-19 for the start year [%].	required
covid_end_year_passenger	Number	Covid-19 end year [yr].	required
covid_end_year_reference_rpk_ratio	float	Percentage of traffic level reached in Covid-19 end year compared with the one in Covid-19 start year [%].	required
cagr_passenger_short_range_reference_periods	list	Reference periods for the CAGR for passenger short-range market [yr].	required
cagr_passenger_short_range_reference_periods_values	list	CAGR for passenger short-range market for the reference periods [%].	required
cagr_passenger_medium_range_reference_periods	list	Reference periods for the CAGR for passenger medium-range market [yr].	required
cagr_passenger_medium_range_reference_periods_values	list	CAGR for passenger medium-range market for the reference periods [%].	required
cagr_passenger_long_range_reference_periods	list	Reference periods for the CAGR for passenger long-range market [yr].	required
cagr_passenger_long_range_reference_periods_values	list	CAGR for passenger long-range market for the reference periods [%].	required
rpk_short_range_measures_impact	Series	Traffic reduction impact of specific measures for passenger short-range market [%].	required
rpk_medium_range_measures_impact	Series	Traffic reduction impact of specific measures for passenger medium-range market [%].	required
rpk_long_range_measures_impact	Series	Traffic reduction impact of specific measures for passenger long-range market [%].	required
airfare_per_rpk	Series	Airfare per RPK [€/RPK].	required
price_elasticity	float	Price elasticity of demand [-].	required

Returns:

Type	Description
rpk_short_range	Number of Revenue Passenger Kilometer (RPK) for passenger short-range market [RPK].
rpk_medium_range	Number of Revenue Passenger Kilometer (RPK) for passenger medium-range market [RPK].
rpk_long_range	Number of Revenue Passenger Kilometer (RPK) for passenger long-range market [RPK].
rpk	Number of Revenue Passenger Kilometer (RPK) for total passenger air transport [RPK].
rpk_no_elasticity	RPKs without considering price elasticity [RPK].
rpk_short_range_no_elasticity	Short-range RPKs without considering price elasticity [RPK].
rpk_medium_range_no_elasticity	Medium-range RPKs without considering price elasticity [RPK].
rpk_long_range_no_elasticity	Long-range RPKs without considering price elasticity [RPK].
annual_growth_rate_passenger_short_range	Annual growth rate for short-range passengers [%/year].
annual_growth_rate_passenger_medium_range	Annual growth rate for medium-range passengers [%/year].
annual_growth_rate_passenger_long_range	Annual growth rate for long-range passengers [%/year].
annual_growth_rate_passenger	Annual growth rate for total passengers [%/year].
cagr_rpk_short_range	Air traffic CAGR over prospective_years for passenger short-range market [%].
cagr_rpk_medium_range	Air traffic CAGR over prospective_years for passenger medium-range market [%].
cagr_rpk_long_range	Air traffic CAGR over prospective_years for passenger long-range market [%].
cagr_rpk	Air traffic CAGR over prospective_years for total passenger market [%].
prospective_evolution_rpk_short_range	Evolution in percentage of Revenue Passenger Kilometer (RPK) for passenger short-range market between prospection_start_year and end_year [%].
prospective_evolution_rpk_medium_range	Evolution in percentage of Revenue Passenger Kilometer (RPK) for passenger medium-range market between prospection_start_year and end_year [%].
prospective_evolution_rpk_long_range	Evolution in percentage of Revenue Passenger Kilometer (RPK) for passenger long-range market between prospection_start_year and end_year [%].
prospective_evolution_rpk	Evolution in percentage of Revenue Passenger Kilometer (RPK) for total passenger market between prospection_start_year and end_year [%].

Source code in aeromaps/models/air_transport/air_traffic/price_elasticity.py

def compute(
    self,
    rpk_init: pd.Series,
    short_range_rpk_share_2019: float,
    medium_range_rpk_share_2019: float,
    long_range_rpk_share_2019: float,
    covid_start_year: Number,
    covid_rpk_drop_start_year: float,
    covid_end_year_passenger: Number,
    covid_end_year_reference_rpk_ratio: float,
    cagr_passenger_short_range_reference_periods: list,
    cagr_passenger_short_range_reference_periods_values: list,
    cagr_passenger_medium_range_reference_periods: list,
    cagr_passenger_medium_range_reference_periods_values: list,
    cagr_passenger_long_range_reference_periods: list,
    cagr_passenger_long_range_reference_periods_values: list,
    rpk_short_range_measures_impact: pd.Series,
    rpk_medium_range_measures_impact: pd.Series,
    rpk_long_range_measures_impact: pd.Series,
    airfare_per_rpk: pd.Series,
    price_elasticity: float,
) -> Tuple[
    pd.Series,
    pd.Series,
    pd.Series,
    pd.Series,
    pd.Series,
    pd.Series,
    pd.Series,
    pd.Series,
    pd.Series,
    pd.Series,
    pd.Series,
    pd.Series,
    float,
    float,
    float,
    float,
    float,
    float,
    float,
    float,
]:
    """
    Execute the computation of prospective air traffic.

    Parameters
    ----------
    rpk_init
        Historical number of Revenue Passenger Kilometer (RPK) over 2000-2019 [RPK].
    short_range_rpk_share_2019
        Share of RPK from short-range market in 2019 [%].
    medium_range_rpk_share_2019
        Share of RPK from medium-range market in 2019 [%].
    long_range_rpk_share_2019
        Share of RPK from long-range market in 2019 [%].
    covid_start_year
        Covid-19 start year [yr].
    covid_rpk_drop_start_year
        Drop in RPK due to Covid-19 for the start year [%].
    covid_end_year_passenger
        Covid-19 end year [yr].
    covid_end_year_reference_rpk_ratio
        Percentage of traffic level reached in Covid-19 end year compared with the one in Covid-19 start year [%].
    cagr_passenger_short_range_reference_periods
        Reference periods for the CAGR for passenger short-range market [yr].
    cagr_passenger_short_range_reference_periods_values
        CAGR for passenger short-range market for the reference periods [%].
    cagr_passenger_medium_range_reference_periods
        Reference periods for the CAGR for passenger medium-range market [yr].
    cagr_passenger_medium_range_reference_periods_values
        CAGR for passenger medium-range market for the reference periods [%].
    cagr_passenger_long_range_reference_periods
        Reference periods for the CAGR for passenger long-range market [yr].
    cagr_passenger_long_range_reference_periods_values
        CAGR for passenger long-range market for the reference periods [%].
    rpk_short_range_measures_impact
        Traffic reduction impact of specific measures for passenger short-range market [%].
    rpk_medium_range_measures_impact
        Traffic reduction impact of specific measures for passenger medium-range market [%].
    rpk_long_range_measures_impact
        Traffic reduction impact of specific measures for passenger long-range market [%].
    airfare_per_rpk
        Airfare per RPK [€/RPK].
    price_elasticity
        Price elasticity of demand [-].

    Returns
    -------
    rpk_short_range
        Number of Revenue Passenger Kilometer (RPK) for passenger short-range market [RPK].
    rpk_medium_range
        Number of Revenue Passenger Kilometer (RPK) for passenger medium-range market [RPK].
    rpk_long_range
        Number of Revenue Passenger Kilometer (RPK) for passenger long-range market [RPK].
    rpk
        Number of Revenue Passenger Kilometer (RPK) for total passenger air transport [RPK].
    rpk_no_elasticity
        RPKs without considering price elasticity [RPK].
    rpk_short_range_no_elasticity
        Short-range RPKs without considering price elasticity [RPK].
    rpk_medium_range_no_elasticity
        Medium-range RPKs without considering price elasticity [RPK].
    rpk_long_range_no_elasticity
        Long-range RPKs without considering price elasticity [RPK].
    annual_growth_rate_passenger_short_range
        Annual growth rate for short-range passengers [%/year].
    annual_growth_rate_passenger_medium_range
        Annual growth rate for medium-range passengers [%/year].
    annual_growth_rate_passenger_long_range
        Annual growth rate for long-range passengers [%/year].
    annual_growth_rate_passenger
        Annual growth rate for total passengers [%/year].
    cagr_rpk_short_range
        Air traffic CAGR over prospective_years for passenger short-range market [%].
    cagr_rpk_medium_range
        Air traffic CAGR over prospective_years for passenger medium-range market [%].
    cagr_rpk_long_range
        Air traffic CAGR over prospective_years for passenger long-range market [%].
    cagr_rpk
        Air traffic CAGR over prospective_years for total passenger market [%].
    prospective_evolution_rpk_short_range
        Evolution in percentage of Revenue Passenger Kilometer (RPK) for passenger short-range market between prospection_start_year and end_year [%].
    prospective_evolution_rpk_medium_range
        Evolution in percentage of Revenue Passenger Kilometer (RPK) for passenger medium-range market between prospection_start_year and end_year [%].
    prospective_evolution_rpk_long_range
        Evolution in percentage of Revenue Passenger Kilometer (RPK) for passenger long-range market between prospection_start_year and end_year [%].
    prospective_evolution_rpk
        Evolution in percentage of Revenue Passenger Kilometer (RPK) for total passenger market between prospection_start_year and end_year [%].
    """

    hist_index = range(self.historic_start_year, self.prospection_start_year)
    covid_years = range(covid_start_year, covid_end_year_passenger + 1)
    proj_years = range(covid_end_year_passenger + 1, self.end_year + 1)
    # all_years = range(self.historic_start_year, self.end_year + 1)

    self.df.loc[hist_index, "rpk_short_range"] = (
        short_range_rpk_share_2019 / 100 * rpk_init.loc[hist_index]
    )
    self.df.loc[hist_index, "rpk_medium_range"] = (
        medium_range_rpk_share_2019 / 100 * rpk_init.loc[hist_index]
    )
    self.df.loc[hist_index, "rpk_long_range"] = (
        long_range_rpk_share_2019 / 100 * rpk_init.loc[hist_index]
    )

    # Covid functions
    reference_years = [covid_start_year, covid_end_year_passenger]
    reference_values_covid = [
        1 - covid_rpk_drop_start_year / 100,
        covid_end_year_reference_rpk_ratio / 100,
    ]
    covid_function = interp1d(reference_years, reference_values_covid, kind="linear")
    covid_factors = pd.Series(
        [float(covid_function(k)) for k in covid_years], index=covid_years, dtype=float
    )

    # CAGR function
    ## Short range
    annual_growth_rate_passenger_short_range_prospective = aeromaps_leveling_function(
        self,
        cagr_passenger_short_range_reference_periods,
        cagr_passenger_short_range_reference_periods_values,
        model_name=self.name,
    )
    self.df.loc[:, "annual_growth_rate_passenger_short_range"] = (
        annual_growth_rate_passenger_short_range_prospective
    )
    ## Medium range
    annual_growth_rate_passenger_medium_range_prospective = aeromaps_leveling_function(
        self,
        cagr_passenger_medium_range_reference_periods,
        cagr_passenger_medium_range_reference_periods_values,
        model_name=self.name,
    )
    self.df.loc[:, "annual_growth_rate_passenger_medium_range"] = (
        annual_growth_rate_passenger_medium_range_prospective
    )
    ## Long range
    annual_growth_rate_passenger_long_range_prospective = aeromaps_leveling_function(
        self,
        cagr_passenger_long_range_reference_periods,
        cagr_passenger_long_range_reference_periods_values,
        model_name=self.name,
    )
    self.df.loc[:, "annual_growth_rate_passenger_long_range"] = (
        annual_growth_rate_passenger_long_range_prospective
    )

    # Covid période vectorisée
    prev_short = self.df.loc[covid_start_year - 1, "rpk_short_range"]
    prev_medium = self.df.loc[covid_start_year - 1, "rpk_medium_range"]
    prev_long = self.df.loc[covid_start_year - 1, "rpk_long_range"]
    self.df.loc[covid_years, "rpk_short_range"] = prev_short * covid_factors
    self.df.loc[covid_years, "rpk_medium_range"] = prev_medium * covid_factors
    self.df.loc[covid_years, "rpk_long_range"] = prev_long * covid_factors

    # Prospective période vectorisée (hors covid)
    # Short
    growth_short = 1 + self.df.loc[proj_years, "annual_growth_rate_passenger_short_range"] / 100
    self.df.loc[proj_years, "rpk_short_range"] = (
        self.df.loc[covid_end_year_passenger, "rpk_short_range"] * growth_short.cumprod()
    )
    # Medium
    growth_medium = (
        1 + self.df.loc[proj_years, "annual_growth_rate_passenger_medium_range"] / 100
    )
    self.df.loc[proj_years, "rpk_medium_range"] = (
        self.df.loc[covid_end_year_passenger, "rpk_medium_range"] * growth_medium.cumprod()
    )
    # Long
    growth_long = 1 + self.df.loc[proj_years, "annual_growth_rate_passenger_long_range"] / 100
    self.df.loc[proj_years, "rpk_long_range"] = (
        self.df.loc[covid_end_year_passenger, "rpk_long_range"] * growth_long.cumprod()
    )

    rpk_short_range_no_elasticity = self.df["rpk_short_range"].copy()
    rpk_medium_range_no_elasticity = self.df["rpk_medium_range"].copy()
    rpk_long_range_no_elasticity = self.df["rpk_long_range"].copy()

    # rpk_no_elasticity
    self.df.loc[hist_index, "rpk_no_elasticity"] = rpk_init.loc[hist_index]
    self.df.loc[self.prospection_start_year : self.end_year, "rpk_no_elasticity"] = (
        self.df.loc[self.prospection_start_year : self.end_year, "rpk_short_range"].fillna(0)
        + self.df.loc[self.prospection_start_year : self.end_year, "rpk_medium_range"].fillna(0)
        + self.df.loc[self.prospection_start_year : self.end_year, "rpk_long_range"].fillna(0)
    )
    rpk_no_elasticity = self.df["rpk_no_elasticity"]

    self.df.loc[self.historic_start_year : covid_end_year_passenger, "rpk"] = (
        rpk_no_elasticity.loc[self.historic_start_year : covid_end_year_passenger]
    )

    airfare_init = 0.09236379319842411

    self.df.loc[covid_end_year_passenger + 1 : self.end_year, "rpk"] = (
        rpk_no_elasticity
        / (airfare_init**price_elasticity)
        * (
            airfare_per_rpk.loc[covid_end_year_passenger + 1 : self.end_year]
            ** price_elasticity
        )
    )

    # Répartition par segment vectorisée
    rpk_short_range = self.df["rpk_short_range"] * self.df["rpk"] / rpk_no_elasticity
    rpk_medium_range = self.df["rpk_medium_range"] * self.df["rpk"] / rpk_no_elasticity
    rpk_long_range = self.df["rpk_long_range"] * self.df["rpk"] / rpk_no_elasticity

    # TODO discuss if that should be considered for surplus destruction. I think so => not inlcluded in rpk_no_elasticity
    rpk_short_range = rpk_short_range * rpk_short_range_measures_impact
    rpk_medium_range = rpk_medium_range * rpk_medium_range_measures_impact
    rpk_long_range = rpk_long_range * rpk_long_range_measures_impact

    self.df.loc[:, "rpk_short_range"] = rpk_short_range
    self.df.loc[:, "rpk_medium_range"] = rpk_medium_range
    self.df.loc[:, "rpk_long_range"] = rpk_long_range

    # Total RPK vectorisé
    self.df.loc[hist_index, "rpk"] = rpk_init.loc[hist_index]
    self.df.loc[self.prospection_start_year : self.end_year, "rpk"] = (
        self.df.loc[self.prospection_start_year : self.end_year, "rpk_short_range"].fillna(0)
        + self.df.loc[self.prospection_start_year : self.end_year, "rpk_medium_range"].fillna(0)
        + self.df.loc[self.prospection_start_year : self.end_year, "rpk_long_range"].fillna(0)
    )
    rpk = self.df["rpk"]

    # Annual growth rates vectorisés
    idx_growth = range(self.historic_start_year + 1, self.end_year + 1)
    self.df.loc[idx_growth, "annual_growth_rate_passenger_short_range"] = (
        self.df["rpk_short_range"].loc[idx_growth].values
        / self.df["rpk_short_range"].shift(1).loc[idx_growth].values
        - 1
    ) * 100
    self.df.loc[idx_growth, "annual_growth_rate_passenger_medium_range"] = (
        self.df["rpk_medium_range"].loc[idx_growth].values
        / self.df["rpk_medium_range"].shift(1).loc[idx_growth].values
        - 1
    ) * 100
    self.df.loc[idx_growth, "annual_growth_rate_passenger_long_range"] = (
        self.df["rpk_long_range"].loc[idx_growth].values
        / self.df["rpk_long_range"].shift(1).loc[idx_growth].values
        - 1
    ) * 100
    self.df.loc[idx_growth, "annual_growth_rate_passenger"] = (
        self.df["rpk"].loc[idx_growth].values / self.df["rpk"].shift(1).loc[idx_growth].values
        - 1
    ) * 100

    annual_growth_rate_passenger_short_range = self.df[
        "annual_growth_rate_passenger_short_range"
    ]
    annual_growth_rate_passenger_medium_range = self.df[
        "annual_growth_rate_passenger_medium_range"
    ]
    annual_growth_rate_passenger_long_range = self.df["annual_growth_rate_passenger_long_range"]
    annual_growth_rate_passenger = self.df["annual_growth_rate_passenger"]

    # Compound Annual Growth Rate (CAGR)
    cagr_rpk_short_range = 100 * (
        (
            self.df.loc[self.end_year, "rpk_short_range"]
            / self.df.loc[self.prospection_start_year - 1, "rpk_short_range"]
        )
        ** (1 / (self.end_year - self.prospection_start_year))
        - 1
    )
    cagr_rpk_medium_range = 100 * (
        (
            self.df.loc[self.end_year, "rpk_medium_range"]
            / self.df.loc[self.prospection_start_year - 1, "rpk_medium_range"]
        )
        ** (1 / (self.end_year - self.prospection_start_year))
        - 1
    )
    cagr_rpk_long_range = 100 * (
        (
            self.df.loc[self.end_year, "rpk_long_range"]
            / self.df.loc[self.prospection_start_year - 1, "rpk_long_range"]
        )
        ** (1 / (self.end_year - self.prospection_start_year))
        - 1
    )
    cagr_rpk = 100 * (
        (
            self.df.loc[self.end_year, "rpk"]
            / self.df.loc[self.prospection_start_year - 1, "rpk"]
        )
        ** (1 / (self.end_year - self.prospection_start_year))
        - 1
    )

    # Prospective evolution of RPK (between prospection_start_year-1 and end_year)
    prospective_evolution_rpk_short_range = 100 * (
        self.df.loc[self.end_year, "rpk_short_range"]
        / self.df.loc[self.prospection_start_year - 1, "rpk_short_range"]
        - 1
    )
    prospective_evolution_rpk_medium_range = 100 * (
        self.df.loc[self.end_year, "rpk_medium_range"]
        / self.df.loc[self.prospection_start_year - 1, "rpk_medium_range"]
        - 1
    )
    prospective_evolution_rpk_long_range = 100 * (
        self.df.loc[self.end_year, "rpk_long_range"]
        / self.df.loc[self.prospection_start_year - 1, "rpk_long_range"]
        - 1
    )
    prospective_evolution_rpk = 100 * (
        self.df.loc[self.end_year, "rpk"] / self.df.loc[self.prospection_start_year - 1, "rpk"]
        - 1
    )

    self.float_outputs["cagr_rpk_short_range"] = cagr_rpk_short_range
    self.float_outputs["cagr_rpk_medium_range"] = cagr_rpk_medium_range
    self.float_outputs["cagr_rpk_long_range"] = cagr_rpk_long_range
    self.float_outputs["cagr_rpk"] = cagr_rpk
    self.float_outputs["prospective_evolution_rpk_short_range"] = (
        prospective_evolution_rpk_short_range
    )
    self.float_outputs["prospective_evolution_rpk_medium_range"] = (
        prospective_evolution_rpk_medium_range
    )
    self.float_outputs["prospective_evolution_rpk_long_range"] = (
        prospective_evolution_rpk_long_range
    )
    self.float_outputs["prospective_evolution_rpk"] = prospective_evolution_rpk

    return (
        rpk_short_range,
        rpk_medium_range,
        rpk_long_range,
        rpk,
        rpk_no_elasticity,
        rpk_short_range_no_elasticity,
        rpk_medium_range_no_elasticity,
        rpk_long_range_no_elasticity,
        annual_growth_rate_passenger_short_range,
        annual_growth_rate_passenger_medium_range,
        annual_growth_rate_passenger_long_range,
        annual_growth_rate_passenger,
        cagr_rpk_short_range,
        cagr_rpk_medium_range,
        cagr_rpk_long_range,
        cagr_rpk,
        prospective_evolution_rpk_short_range,
        prospective_evolution_rpk_medium_range,
        prospective_evolution_rpk_long_range,
        prospective_evolution_rpk,
    )