wolfhece 2.2.38__py3-none-any.whl → 2.2.39__py3-none-any.whl

wolfhece/MulticriteriAnalysis.py

@@ -0,0 +1,1579 @@
+"""
+This module contains the objects needed to perform
+a multi-criteria analysis for fish passage
+on spatially distributed hydrodynamic data.
+
+The module tests are located in the folder:
+tests/test_MulticriteriAnalysis.py file,
+in HECEPython repository.
+
+Authors: Utashi Ciraane,
+HECE, University of Liège, Belgium.
+"""
+# **************************************************************
+# Libraries used throughout the module
+# **************************************************************
+
+# To stay inside the HECEPython repository
+# ----------------------------------------
+
+# Standard libraries
+# -------------------
+import logging
+import numpy as np
+import os
+
+# Cherry-picked objects from standard libraries
+# ---------------------------------------------
+from enum import Enum
+from pathlib import Path
+from tqdm import tqdm
+from typing import Literal, Union
+
+# Cherry-picked objects from wolfhece
+# ------------------------------------
+from wolfhece.wolf_array import WolfArray, header_wolf
+
+## **************************************************************
+# Constants Used
+# **************************************************************
+class Operator(Enum):
+    """
+    Mathematical operators used
+    in the module.
+    """
+    AVERAGE = "average"
+    BETWEEN = "between" # Between two limits. The limits are included.
+    BETWEEN_STRICT = "strictly between" # Between two limits. The limits are not included.
+    EQUAL = "equal"  # Equal to a value.
+    INFERIOR = "inferior" # Less than a limit. The limit is not included.
+    INFERIOR_OR_EQUAL = "inferior or equal" # Less than or equal to a limit. The limit is included.
+    OUTSIDE = "outside" # Outside two limits. The limits are included.
+    OUTSIDE_STRICT = "strictly outside" # Outside two limits. The limits are not included.
+    PERCENTAGE = "percentage"
+    PRODUCT = "product"
+    SUM = "sum"
+    SUPERIOR = "superior" # Greater than a limit. The limit is not included.
+    SUPERIOR_OR_EQUAL = "superior or equal" # Greater than or equal to a limit. The limit is included.
+    THRESHOLD = "threshold"
+    WEIGHTED_SUM = "weighted sum"
+
+class Format(Enum):
+    """
+    Types of data formats used in the module.
+    """
+    FLOAT32 = np.float32  # The float32 data type used in the module.
+    FLOAT64 = np.float64  # The float64 data type used in the module.
+    INT32 = np.int32  # The int32 data type used in the module.
+    INT64 = np.int64  # The int64 data type used in the module.
+
+class Status(Enum):
+    """
+    Status of objects in the module
+    """
+    PROGRESS_BAR_DISABLE = False    # Enable or disable the tqdm  progress bars.
+
+
+class Constant(Enum):
+    """
+    This class contains the constants used throughout the module.
+    """
+    EPSG = 31370
+    ONE = 1
+    ZERO = 0
+    CONDITION = "condition"
+
+
+## **************************************************************
+# Classes or Objects used to perform the multi-criteria analysis
+## **************************************************************
+class Variable:
+    """
+    This class is used to store variable (data: array)
+    that will be used  in the multi-criteria analysis.
+
+    The types of variable allowed are:
+        - np.ma.MaskedArray: A numpy masked array.
+        - np.ndarray: A numpy array.
+        - WolfArray: A WolfArray object from the wolfhece library.
+    """
+    def __init__(self,
+                 variable: np.ma.MaskedArray| np.ndarray| WolfArray,
+                 dtype : Union[np.dtype, str] = Format.FLOAT32.value) :
+        """
+        Initialize the variable with a WolfArray or a numpy array.
+        if a type dtype is not provided, it defaults to np.float32.
+
+        :param variable: A WolfArray or a numpy array.
+        ::type variable: np.ma.MaskedArray | np.ndarray | WolfArray
+        :param dtype: The data type of the variable, defaults to np.float32.
+        ::type dtype: Union[np.dtype, str]
+
+        """
+        self._variable: np.ma.MaskedArray = None
+        self._dtype = dtype
+        self.variable = variable
+
+
+
+    @property
+    def variable(self) -> np.ma.MaskedArray:
+        """
+        Return the variable.
+
+        :return: The variable as a numpy masked array.
+        """
+        return self._variable
+
+    @variable.setter
+    def variable(self, value: np.ma.MaskedArray| np.ndarray| WolfArray) :
+        """
+        Set the variable and converts it to a numpy masked array.
+
+        :param value: A WolfArray or a numpy masked array.
+        """
+        assert isinstance(value, (np.ma.MaskedArray, np.ndarray, WolfArray)), \
+            "The variable must be a numpy masked array, a numpy array or a WolfArray."
+        if isinstance(value, WolfArray):
+            if np.ma.is_masked(value.array):
+                value = value.array
+            else:
+                # If the WolfArray is not masked, create a masked array with no mask.
+                mask = np.zeros(value.array.shape, dtype=bool)
+                value = np.ma.masked_array(value.array, mask=mask)
+
+        elif isinstance(value, np.ndarray):
+            mask =np.zeros(value.shape, dtype=bool)
+            value = np.ma.masked_array(value, mask=mask)
+        self._variable = value.astype(self._dtype, copy=True)
+
+    @property
+    def dtype(self) -> type:
+        """
+        Return the data type of the variable.
+
+        :return: The data type of the variable.
+        """
+        return self._variable.dtype
+
+    @dtype.setter
+    def dtype(self, value: Union[type, str]) -> None:
+        """
+        Set the data type of the variable.
+
+        :param value: The data type to set.
+        """
+        assert isinstance(value, (type, str)), \
+            "The data type must be a numpy dtype or a string representing a numpy dtype."
+        # if isinstance(value, str):
+        #     # value = type(value)
+
+        self._variable = self._variable.astype(value, copy=True)
+        self._dtype = value
+
+class Criteria:
+    """
+    This class is used to define a criteria included in the multi-criteria analysis
+
+    It contains a threshold and a condition:.
+        - superior,
+        - inferior,
+        - equal,
+        - superior or equal,
+        - inferior or equal,
+        - between,
+        - strictly between,
+        - outside,
+        - strictly outside.
+
+    The condition defines how the threshold is applied to select
+    values in the variable true (1) or False (0).
+    """
+    def __init__(self,
+                 threshold: Union[float, int, tuple],
+                 condition: Literal[Operator.SUPERIOR.value,
+                                   Operator.INFERIOR.value,
+                                   Operator.SUPERIOR_OR_EQUAL.value,
+                                   Operator.INFERIOR_OR_EQUAL.value,
+                                   Operator.BETWEEN.value,
+                                   Operator.BETWEEN_STRICT.value,
+                                   Operator.OUTSIDE.value,
+                                   Operator.OUTSIDE_STRICT.value
+                                   ] = Operator.SUPERIOR_OR_EQUAL.value) -> dict:
+        """
+        Return a criteria with a threshold and a condition.
+
+        :param threshold: The threshold value for the criteria.
+        :type threshold: float| int|tuple
+        :param condition: The condition to select the data around the threshold.
+        :type condition: str
+        :return: a dictionary containing 2 elements, the threshold and the condition.
+        :rtype: dict
+        """
+
+        self._available_conditions = [Operator.SUPERIOR.value,
+                                   Operator.INFERIOR.value,
+                                   Operator.EQUAL.value,
+                                   Operator.SUPERIOR_OR_EQUAL.value,
+                                   Operator.INFERIOR_OR_EQUAL.value,
+                                   Operator.BETWEEN.value,
+                                   Operator.BETWEEN_STRICT.value,
+                                   Operator.OUTSIDE.value,
+                                   Operator.OUTSIDE_STRICT.value
+                                   ] # add NOT in method too.
+        self._needs_tuple = (Operator.BETWEEN.value,
+                        Operator.BETWEEN_STRICT.value,
+                        Operator.OUTSIDE.value,
+                        Operator.OUTSIDE_STRICT.value)
+        self._criteria: dict = {}
+
+        self.threshold = threshold
+        self.condition = condition
+
+    @property
+    def criteria(self) -> dict:
+        """
+        Return the criteria.
+
+        :return: The criteria as a dictionary.
+        """
+        return self._criteria
+
+    @criteria.setter
+    def criteria(self,
+                 criteria : dict) -> None:
+        """
+        Set the criteria.
+
+        :param value: The criteria as a dictionary.
+        """
+        assert isinstance(criteria, dict), f"The criteria must be a dictionary, not {type(criteria)}."
+        assert Operator.THRESHOLD.value in criteria,\
+            f"The threshold key must be {Operator.THRESHOLD.value } in the criteria dictionary."
+        assert Constant.CONDITION.value in criteria,\
+            f"The condition key must be {Constant.CONDITION.value} in the criteria dictionary."
+        self.threshold = criteria[Operator.THRESHOLD.value]
+        self.condition = criteria[Constant.CONDITION.value]
+
+    @property
+    def threshold(self) -> float:
+        """
+        Get the threshold of the criteria.
+
+        :return: The threshold as a float.
+        """
+        return self._criteria[Operator.THRESHOLD.value]
+
+    @threshold.setter
+    def threshold(self, value: float) -> None:
+        """
+        Set the threshold of the criteria.
+
+        :param value: The threshold to set.
+        """
+        assert isinstance(value, (float, int, tuple)),\
+              f"The threshold must be a float or an int or a tuple of length equals to Two, not a {type(value)} ."
+        if isinstance(value, tuple):
+            assert len(value) == 2, "The threshold tuple must have exactly two elements."
+            assert value[0] <= value[1], "The first element of the threshold tuple must be less than the second element."
+            if Constant.CONDITION.value in self._criteria:
+                if self.condition not in (self._needs_tuple):
+                    value = self._tuple_to_number(value)
+                    logging.warning(f"The condition '{self.condition}' does not require a tuple for the threshold. "
+                                    f"The threshold will be set to : {value} which i the first value of the tuple.")
+
+        elif isinstance(value, (float, int)):
+            if Constant.CONDITION.value in self._criteria:
+                if self.condition in self._needs_tuple:
+                    logging.warning(f"The condition '{self.condition}' requires a tuple for the threshold. "
+                                    f"The threshold will be set to a tuple with the value {value} as both elements.")
+                    # Convert single number to tuple
+                    value = self._number_to_tuple(value)
+        assert isinstance(self._criteria, dict), f"The criteria type must be a dictionary, not {type(self._criteria)}."
+        self._criteria[Operator.THRESHOLD.value] = value
+
+    @property
+    def condition(self) -> str:
+        """
+        Get the condition.
+
+        :return: The condition as a string.
+        """
+        return self._criteria[Constant.CONDITION.value]
+
+    @condition.setter
+    def condition(self, value: Literal[Operator.SUPERIOR.value,
+                                        Operator.INFERIOR.value,
+                                        Operator.EQUAL.value,
+                                        Operator.SUPERIOR_OR_EQUAL.value,
+                                        Operator.INFERIOR_OR_EQUAL.value,
+                                        Operator.BETWEEN.value,
+                                        Operator.BETWEEN_STRICT.value
+                                        ]) -> None:
+        """
+        Set the condition.
+
+        :param value: The condition to set.
+        """
+        assert isinstance(value, str), f"The condition must be a string, not a {type(value)}."
+        assert value in self._available_conditions,\
+            f"The criteria must be one of the following: {self._available_conditions}."
+        self._criteria[Constant.CONDITION.value] = value
+        # If the condition is a range condition (between, strictly between, outside, strictly outside),
+        # we need to convert the threshold to a tuple if it is not already.
+        if value in self._needs_tuple:
+            if isinstance(self.threshold, (int, float)):
+                # If the threshold is a single number, convert it to a tuple
+                self.threshold = self._number_to_tuple(self.threshold)
+        else:
+            # If the condition does not require a tuple, ensure the threshold is a single number
+            if isinstance(self.threshold, tuple):
+                self.threshold = self._tuple_to_number(self.threshold)
+
+    def _number_to_tuple(self, value: Union[float, int]) -> tuple:
+        """
+        Convert a number to a tuple of two elements.
+
+        :param value: The number to convert.
+        :return: A tuple with the number as both elements.
+        """
+        assert isinstance(value, (float, int)), f"The value must be a float or an int, not {type(value)}."
+        return (value, value)
+
+    def _tuple_to_number(self, value: tuple) -> float:
+        """
+        Convert a tuple of two elements to a number.
+
+        :param value: The tuple to convert.
+        :return: The first element of the tuple as a float.
+        """
+        assert isinstance(value, tuple), f"The value must be a tuple, not {type(value)}."
+        assert len(value) == 2, "The tuple must have exactly two elements."
+        return value[0]
+
+class Score:
+    """
+    This class  scores a Variable based on the given criteria (threshold & condition).
+
+    The score is a np.MaskedArray with the same shape as the given variable.
+
+    The score contains boolean values indicating whether the variable meets the criteria (1) or (0).
+    If the variable is masked, the score is also masked.
+
+    Binary notation was selected to allow mathematical operations.
+    """
+    def __init__(self,
+                 variable: Variable,
+                 criteria: Criteria,
+                 dtype = Format.INT32.value
+                 ) -> None:
+        """
+        Initialize the score with a variable and a criteria.
+
+        :param variable: The variable (array) to score.
+        :type variable: Variable
+        :param criteria: The criteria to use for scoring (threshold and condition).
+        :type criteria: Criteria
+        :param dtype: The data type of the score, defaults to np.int32.
+        :type dtype: Union[np.dtype, str]
+        """
+        assert isinstance(variable, Variable), f"The variable must be an instance of Variable, not {type(variable)}."
+        assert isinstance(criteria, Criteria), f"The criteria must be an instance of Criteria, not {type(criteria)}."
+        self._variable = None
+        self.variable = variable
+        self._criteria = None
+        self.criteria = criteria
+        self._score: np.ma.MaskedArray = None
+        self._dtype = None
+        self.dtype = dtype
+
+    @property
+    def variable(self) -> Variable:
+        """
+        Get the variable.
+
+        :return: The variable as a Variable object.
+        """
+        return self._variable
+
+    @variable.setter
+    def variable(self, value: Variable) -> None:
+        """
+        Set the variable.
+
+        :param value: The variable to set.
+        """
+        assert isinstance(value, Variable), f"The variable must be an instance of Variable, not {type(value)}."
+        self._variable = value
+
+    @property
+    def criteria(self) -> Criteria:
+        """
+        Get the criteria.
+
+        :return: The criteria as a Criteria object.
+        """
+        return self._criteria
+
+    @criteria.setter
+    def criteria(self, value: Criteria) -> None:
+        """
+        Set the criteria.
+
+        :param value: The criteria to set.
+        """
+        assert isinstance(value, Criteria), f"The criteria must be an instance of Criteria, not {type(value)}."
+        self._criteria = value
+
+    @property
+    def score(self) -> np.ma.MaskedArray:
+        """
+        Get the score.
+
+        :return: The score as a numpy masked array.
+        """
+        self.score = self._compute_score()
+        return self._score
+
+    @score.setter
+    def score(self, value: np.ma.MaskedArray) -> None:
+        """
+        Set the score.
+
+        :param value: The score to set.
+        """
+        assert isinstance(value, np.ma.MaskedArray), f"The score must be a numpy masked array, not {type(value)}."
+        assert value.shape == self.variable.variable.shape, \
+            f"The score must have the same shape as the variable, not {value.shape}."
+        self._score = value.astype(self._dtype, copy=True)
+
+    @property
+    def dtype(self) -> type:
+        """
+        Get the data type of the score.
+
+        :return: The data type of the score.
+        """
+        return self._dtype
+
+    @dtype.setter
+    def dtype(self, value: Union[type, str]) -> None:
+        """
+        Set the data type of the score.
+
+        :param value: The data type to set.
+        """
+        assert isinstance(value, (type, str)), f"The data type must be a numpy dtype or a string representing a numpy dtype, not {type(value)}."
+        assert np.issubdtype(value, np.integer), \
+            f"The data type must be an integer type, not {value}."
+        self._dtype = value
+        if self._score is not None:
+            self._score = self._score.astype(value, copy=True)
+
+    def _compute_score(self):
+        """
+        Compute the score based on the variable and criteria.
+
+        :return: The score as a numpy masked array.
+        """
+        assert self._variable is not None, "Variable is not set."
+        assert self._criteria is not None, "Criteria is not set."
+        assert isinstance(self._variable.variable, np.ma.MaskedArray), \
+            f"The variable must be a numpy masked array, not {type(self._variable.variable)}"
+        score:np.ma.MaskedArray = self.variable.variable.copy()
+        score = score.astype(Format.INT32.value, copy=True)  # Convert to int32 for scoring
+
+        if self.criteria.condition == Operator.SUPERIOR.value:
+            score[self.variable.variable <= self.criteria.threshold] = Constant.ZERO.value
+            score[self.variable.variable > self.criteria.threshold] = Constant.ONE.value
+
+        elif self.criteria.condition == Operator.INFERIOR.value:
+            score[self.variable.variable >= self.criteria.threshold] = Constant.ZERO.value
+            score[self.variable.variable < self.criteria.threshold] = Constant.ONE.value
+
+        elif self.criteria.condition == Operator.EQUAL.value:
+            score[self.variable.variable != self.criteria.threshold] = Constant.ZERO.value
+            score[self.variable.variable == self.criteria.threshold] = Constant.ONE.value
+
+        elif self.criteria.condition == Operator.SUPERIOR_OR_EQUAL.value:
+            score[self.variable.variable < self.criteria.threshold] = Constant.ZERO.value
+            score[self.variable.variable >= self.criteria.threshold] = Constant.ONE.value
+
+        elif self.criteria.condition == Operator.INFERIOR_OR_EQUAL.value:
+            score[self.variable.variable > self.criteria.threshold] = Constant.ZERO.value
+            score[self.variable.variable <= self.criteria.threshold] = Constant.ONE.value
+
+        elif self.criteria.condition == Operator.BETWEEN.value:
+            assert isinstance(self.criteria.threshold, tuple), \
+                "The threshold for the 'between' condition must be a tuple of two elements."
+            score[(self.variable.variable < self.criteria.threshold[0]) | \
+                  (self.variable.variable > self.criteria.threshold[1])] = Constant.ZERO.value
+            score[(self.variable.variable >= self.criteria.threshold[0]) & \
+                  (self.variable.variable <= self.criteria.threshold[1])] = Constant.ONE.value
+
+        elif self.criteria.condition == Operator.BETWEEN_STRICT.value:
+            assert isinstance(self.criteria.threshold, tuple), \
+                "The threshold for the 'strictly between' condition must be a tuple of two elements."
+            score[(self.variable.variable <= self.criteria.threshold[0]) | \
+                  (self.variable.variable >= self.criteria.threshold[1])] = Constant.ZERO.value
+            score[(self.variable.variable > self.criteria.threshold[0]) & \
+                  (self.variable.variable < self.criteria.threshold[1])] = Constant.ONE.value
+
+        elif self.criteria.condition == Operator.OUTSIDE.value:
+            assert isinstance(self.criteria.threshold, tuple), \
+                "The threshold for the 'outside' condition must be a tuple of two elements."
+            score[(self.variable.variable > self.criteria.threshold[0]) & \
+                  (self.variable.variable < self.criteria.threshold[1])] = Constant.ZERO.value
+            score[(self.variable.variable <= self.criteria.threshold[0]) | \
+                  (self.variable.variable >= self.criteria.threshold[1])] = Constant.ONE.value
+
+        elif self.criteria.condition == Operator.OUTSIDE_STRICT.value:
+            assert isinstance(self.criteria.threshold, tuple), \
+                "The threshold for the 'strictly outside' condition must be a tuple of two elements."
+            score[(self.variable.variable >= self.criteria.threshold[0]) | \
+                  (self.variable.variable <= self.criteria.threshold[1])] = Constant.ZERO.value
+            score[(self.variable.variable < self.criteria.threshold[0]) | \
+                  (self.variable.variable > self.criteria.threshold[1])] = Constant.ONE.value
+
+        else:
+            raise ValueError(f"Unknown condition: {self._criteria.condition}. "
+                             f"Available conditions are: {self._criteria._available_conditions}")
+
+        return score.astype(self.dtype, copy=True)
+
+class Scores:
+    """
+    This class is used to store multiple scores from different criteria as dictionary.
+
+    It contains a dictionary of scores,
+    where the keys are the score names and the values are the Score objects.
+    """
+    def __init__(self, scores:dict[str, Score]) -> None:
+        """
+        Initialize the Scores object.
+        """
+        self._scores: dict[str, Score] = None
+        self.scores = scores
+
+    @property
+    def scores(self) -> dict[str, Score]:
+        """
+        Get the scores.
+
+        :return: The scores as a dictionary of Score objects.
+        """
+        return self._scores
+
+    @scores.setter
+    def scores(self, value: dict[str, Score]) -> None:
+        """
+        Set the scores.
+
+        :param value: The scores to set.
+        """
+        assert isinstance(value, dict), f"The scores must be a dictionary, not {type(value)}."
+        assert all(isinstance(key, str) for key in value.keys()), \
+            "All keys in the scores dictionary must be strings."
+        assert all(isinstance(score, Score) for score in value.values()), \
+            "All values in the scores dictionary must be Score objects."
+        assert all(score.variable.variable.shape == next(iter(value.values())).variable.variable.shape \
+                    for score in value.values()), "All scores must have the same shape."
+        self._scores = value
+
+    @property
+    def number(self) -> int:
+        """
+        Get the number of scores.
+
+        :return: The number of scores in the dictionary.
+        """
+        return len(self._scores) if self._scores is not None else 0
+
+    def add_score(self, name: str, score: Score) -> None:
+        """
+        Add a score to the scores dictionary.
+
+        :param name: The name of the score.
+        :param score: The Score object to add.
+        """
+        assert isinstance(name, str), f"The name must be a string, not {type(name)}."
+        assert isinstance(score, Score), f"The score must be an instance of Score, not {type(score)}."
+        if self._scores is None:
+            self._scores = {}
+        self._scores[name] = score
+
+    def get_score(self, name: str) -> Score:
+        """
+        Get a score by its name.
+
+        :param name: The name of the score to get.
+        :return: The Score object with the given name.
+        """
+        try:
+            return self._scores[name]
+        except KeyError:
+            logging.error(f"The score with name: '{name}' does not exist.")
+            raise KeyError(f"The score with name '{name}' does not exist.")
+
+    def remove_score(self, name: str) -> None:
+        """
+        Remove a score by its name.
+
+        :param name: The name of the score to remove.
+        """
+        try:
+            del self._scores[name]
+        except KeyError:
+            logging.error(f"The score with name:'{name}' does not exist.")
+            raise KeyError(f"The score with name '{name}' does not exist.")
+
+class Operations:
+    """
+    This class is used to perform mathematical operations on scores.
+
+    N.B.: if needed operations from other libraries can be added here.
+    for instance **pymcdm, scikit-mcda, etc.**
+    """
+
+    def __init__(self,
+                 scores: Scores,
+                 int_type:str = Format.INT32.value,
+                 float_type:str = Format.FLOAT32.value,
+                 weight: dict = None
+                ) -> None:
+        """
+        Initialize the Operations object with a Scores object.
+
+        :param scores: The Scores object to perform operations on.
+        :type scores: Scores
+        :param int_type: The integer data type to use in the operations if integers are expected,
+        defaults to np.int32.
+        :type int_type: str
+        :param float_type: The float data type to use in the operations if floats are expected
+        , defaults to np.float32.
+        :type float_type: str
+        :param weight: A dictionary of weights for scores, defaults to None.
+        :type weight: dict
+
+        :raises AssertionError: If the scores is not an instance of Scores.
+        """
+        assert isinstance(scores, Scores),\
+            f"The scores must be an instance of Scores, not {type(scores)}."
+        self._scores = None
+        self._int_type = None
+        self._float_type = None
+        self._weight = None
+        self.scores = scores
+        self.int_type = int_type
+        self.float_type = float_type
+        self.weight = weight
+        self._available_operations = [Operator.SUM.value,
+                                      Operator.PRODUCT.value,
+                                      Operator.WEIGHTED_SUM.value,
+                                      Operator.AVERAGE.value]
+
+    @property
+    def scores(self) -> Scores:
+        """
+        Get the Scores object.
+
+        :return: The Scores object.
+        """
+        return self._scores
+
+    @scores.setter
+    def scores(self, value: Scores) -> None:
+        """
+        Set the Scores object.
+
+        :param value: The Scores object to set.
+        """
+        assert isinstance(value, Scores),\
+            f"The scores must be an instance of Scores, not {type(value)}."
+        self._scores = value
+
+    @property
+    def int_type(self) -> str:
+        """
+        Get the integer data type used in the operations.
+
+        :return: The integer data type as a string.
+        """
+        return self._int_type
+    @int_type.setter
+    def int_type(self, value: Union[str,type]) -> None:
+        """
+        Set the integer data type used in the operations.
+
+        :param value: The integer data type to set.
+        """
+        assert isinstance(value, (str,type)), f"The integer type must be a string, not {type(value)}."
+        self._int_type = value
+
+    @property
+    def float_type(self) -> str:
+        """
+        Get the float data type used in the operations.
+
+        :return: The float data type as a string.
+        """
+        return self._float_type
+
+    @float_type.setter
+    def float_type(self, value: Union[str,type]) -> None:
+        """
+        Set the float data type used in the operations.
+
+        :param value: The float data type to set.
+        """
+        assert isinstance(value, (str,type)), f"The float type must be a string, not {type(value)}."
+        self._float_type = value
+
+    @property
+    def weight(self) -> dict:
+        """
+        Get the weight dictionary.
+
+        :return: The weight dictionary.
+        """
+        return self._weight
+
+    @weight.setter
+    def weight(self, value: dict) -> None:
+        """
+        Set the weight dictionary.
+
+        :param value: The weight dictionary to set.
+        """
+        if value is None:
+            if self.scores.number == 0:
+                logging.warning("No scores to set weight for. Setting weight to None.")
+                self._weight = None
+                return
+            elif self.scores.number > 0:
+                logging.warning("No weight provided. Setting weight to equal distribution.")
+                value = {key: 1  for key in self.scores.scores.keys()}
+        else:
+            assert isinstance(value, dict), f"The weight must be a dictionary, not {type(value)}."
+            assert len(value) == self.scores.number,\
+                "The weight dictionary must have the same number of keys as the scores dictionary."
+            assert all(key in self.scores.scores.keys() for key in value.keys()),\
+                "All keys in the weight dictionary must be present in the scores dictionary."
+        self._weight = value
+
+    def sum(self) -> np.ma.MaskedArray:
+        """
+        Sum all scores in the Scores object.
+
+        :return: The sum of all scores as a numpy masked array.
+        """
+        if self.scores.number == 0:
+            logging.warning("No scores to sum.")
+            return np.ma.masked_array([], mask=True)
+        score_sum = np.ma.masked_array(np.zeros_like(next(iter(self.scores.scores.values())).score),
+                                       dtype=Format.INT32.value)
+
+        for score in self.scores.scores.values():
+            score_sum += score.score
+
+        return score_sum
+
+    def product(self) -> np.ma.MaskedArray:
+        """
+        Calculate the product of all scores in the Scores object.
+
+        :return: The product of all scores as a numpy masked array.
+        """
+        if self.scores.number == 0:
+            logging.warning("No scores to multiply.")
+            return np.ma.masked_array([], mask=True)
+
+        score_product = np.ma.masked_array(np.ones_like(next(iter(self.scores.scores.values())).score),
+                                           dtype=Format.INT32.value)
+
+        for score in self.scores.scores.values():
+            score_product *= score.score
+
+        return score_product
+
+
+    def weighted_sum(self) -> np.ma.MaskedArray:
+        """
+        Calculate the weighted sum of all scores in the Scores object.
+
+        :return: The weighted sum of all scores as a numpy masked array.
+        """
+        if self.scores.number == 0:
+            logging.warning("No scores to calculate weighted sum.")
+            return np.ma.masked_array([], mask=True)
+
+        score_sum = np.ma.masked_array(np.zeros_like(next(iter(self.scores.scores.values())).score),
+                                       dtype=self.float_type) # FIXME Find a better method to apply this
+
+        for key, score in self.scores.scores.items():
+            if key in self.weight:
+                array = (score.score * self.weight[key]).astype(self.float_type, copy=True)
+                score_sum = score_sum.astype(self.float_type, copy=True)
+                score_sum += array
+
+            else:
+                raise KeyError(f"No weight provided for score '{key}'. Using default weight of 1.")
+
+        return score_sum
+
+    def average(self) -> np.ma.MaskedArray:
+        """
+        Calculate the average of all scores in the Scores object.
+
+        :return: The average of all scores as a numpy masked array.
+        """
+        if self.scores.number == 0:
+            logging.warning("No scores to average.")
+            return np.ma.masked_array([], mask=True)
+
+        score_sum = self.sum()
+        average: np.ma.MaskedArray = (score_sum / self.scores.number)
+        average = average.astype(self.float_type)
+
+        return average
+
+    def percentage(self) -> np.ma.MaskedArray:
+        """
+        Calculate the percentage of each score in the Scores object.
+
+        :return: The percentage of each score as a numpy masked array.
+        """
+        return self.average() * 100
+
+    def apply_operation(self,
+                        operation: Literal[Operator.SUM.value,
+                                           Operator.PRODUCT.value,
+                                           Operator.WEIGHTED_SUM.value,
+                                           Operator.AVERAGE.value,
+                                           Operator.PERCENTAGE.value
+                                           ]) -> np.ma.MaskedArray:
+        """
+        Select an operation to perform on the scores.
+
+        :param operation: The operation to perform.
+        :return: The result of the operation as a numpy masked array.
+        """
+        # if operation.lower() not in self._available_operations():
+        if operation == Operator.SUM.value:
+            return self.sum()
+        elif operation == Operator.PRODUCT.value:
+            return self.product()
+        elif operation == Operator.WEIGHTED_SUM.value:
+            return self.weighted_sum()
+        elif operation == Operator.AVERAGE.value:
+            return self.average()
+        elif operation == Operator.PERCENTAGE.value:
+            return self.percentage()
+        else:
+            raise ValueError(f"Unknown operation: {operation}. "
+                             f"Available operations are: {self._available_operations()}")
+
+class Results:
+    """
+    This class is used to collect results of the Operations module.
+
+    It allows:
+        - to define the mold (a WolfArray) which serve as the geo-spatial extent computed results,
+        - to define the method used to compute scores,
+        - to get and write the results.
+    """
+    def __init__(self,
+                 operations: Operations,
+                 mold: WolfArray = None,
+                 method: Literal[Operator.SUM.value,
+                                 Operator.PRODUCT.value,
+                                    Operator.WEIGHTED_SUM.value,
+                                    Operator.AVERAGE.value,
+                                    Operator.PERCENTAGE.value
+                                    ] = Operator.SUM.value
+                ) -> None:
+
+        """
+        Initialize the Results object.
+
+        :param operations: The Operations object to perform operations on.
+        :type operations: Operations
+        :param mold: The mold used for the results, defaults to None.
+        :type mold: WolfArray
+        :param method: The method used to compute the scores, defaults sum.
+        :type method: str,
+
+        """
+        self._operations = None
+        self._mold = None
+        self._method = None
+        self.operations = operations
+        self.mold = mold
+        self.method = method
+
+    @property
+    def operations(self) -> Operations:
+        """
+        Get the Operations object.
+
+        :return: The Operations object.
+        """
+        return self._operations
+
+    @operations.setter
+    def operations(self, value: Operations) -> None:
+        """
+        Set the Operations object.
+
+        :param value: The Operations object to set.
+        """
+        assert isinstance(value, Operations),\
+            f"The operations must be an instance of Operations, not {type(value)}."
+        self._operations = value
+
+    @property
+    def mold(self) -> WolfArray:
+        """
+        Get the mold used for the results.
+
+        :return: The mold as a WolfArray.
+        """
+
+        return self._mold
+
+    @mold.setter
+    def mold(self, value: WolfArray) -> None:
+        """
+        Set the mold used for the results.
+        :param value: The mold to set.
+        """
+        if value is not None:
+            assert isinstance(value, WolfArray), f"The mold must be a WolfArray, not {type(value)}."
+            for key in self.operations.scores.scores.keys():
+                if value.shape != self.operations.scores.get_score(key).score.shape:
+                    raise ValueError(f"The mold shape {value.shape} does not match the score shape {self.operations.scores.get_score(key).score.shape}.")
+        self._mold = value
+
+    @property
+    def header(self) -> header_wolf:
+        """
+        Get the header of the mold.
+
+        :return: The header of the mold as a header_wolf object.
+        """
+        if self.mold is None:
+            if self.operations is not None:
+                    sample = self.operations.scores.get_score(next(iter(self.operations.scores.scores.keys()))).score
+                    nbx = sample.shape[0]
+                    nby = sample.shape[1]
+                    return self.create_header_wolf(nbx=nbx,
+                                                        nby=nby,
+                                                        dx= Constant.ONE.value,
+                                                        dy= Constant.ONE.value)
+        else:
+            return self.mold.get_header()
+
+    @property
+    def method(self) -> str:
+        """
+        Get the method used for the results.
+
+        :return: The method as a string.
+        """
+        return self._method
+
+    @method.setter
+    def method(self, value: Literal[Operator.SUM.value,
+                                      Operator.PRODUCT.value,
+                                      Operator.WEIGHTED_SUM.value,
+                                      Operator.AVERAGE.value,
+                                      Operator.PERCENTAGE.value
+                                      ]) -> None:
+        """
+        Set the method used for the results.
+        :param value: The method to set.
+
+        """
+        assert value in self.operations._available_operations,\
+            f"The method must be one of the following: {self.operations._available_operations}."
+        self._method = value
+
+    @property
+    def as_numpy_array(self) -> np.ma.MaskedArray:
+        return self.operations.apply_operation(self.method)
+
+    @property
+    def as_WolfArray(self) -> WolfArray:
+        """
+        Get the results as a WolfArray.
+
+        :return: The results as a WolfArray.
+        """
+        return self.as_results(self.as_numpy_array)
+
+    #------------
+    # Methods
+    #------------
+
+    def create_header_wolf(self,
+                       origx:float = None,
+                       origy: float =None ,
+                       origz: float = None,
+                       dx: float = None,
+                       dy: float = None,
+                       dz: float = None,
+                       nbx: int = None,
+                       nby: int = None,
+                            nbz: int = None):
+        """
+        Create a header_wolf object.
+        The header_wolf object is used to describe the spatial
+        characteristics of the array.
+
+
+        :param origx: The x-coordinate of the origin (in 2D - lower left corner),
+        defaults to None.
+        :type origx: float, optional
+        :param origy: The y-coordinate of the origin (in 2D - lower left corner),
+        defaults to None.
+        :type origy: float, optional
+        :param origz: The z-coordinate of the origin, defaults to None.
+        :type origz: float, optional
+        :param dx: The x-spacing (discretization in the x direction), defaults to None.
+        :type dx: float, optional
+        :param dy: The y-spacing (discretization in the x direction), defaults to None.
+        :type dy: float, optional
+        :param dz: The z-spacing, defaults to None.
+        :type dz: float, optional
+        :param nbx: The number of columns, defaults to None.
+        :type nbx: int, optional
+        :param nby: The number of rows, defaults to None.
+        :type nby: int, optional
+        :param nbz: The number of layers, defaults to None.
+        :type nbz: int, optional
+        :return: A header_wolf object with the given parameters.
+        :rtype: header_wolf
+        """
+        header = header_wolf()
+        if origx is not None:
+            header.origx = origx
+        if origy is not None:
+            header.origy = origy
+        if origz is not None:
+            header.origz = origz
+        if dx is not None:
+            header.dx = dx
+        if dy is not None:
+            header.dy = dy
+        if dz is not None:
+            header.dz = dz
+        if nbx is not None:
+            header.nbx = nbx
+        if nby is not None:
+            header.nby = nby
+        if nbz is not None:
+            header.nbz = nbz
+        return header
+
+    def create_WolfArray(self,
+                        dtype = None) -> WolfArray:
+        """
+        Create an empty WolfArray with the given name, and header.
+
+        :param dtype: The data type of the WolfArray, defaults to np.float32.
+        :type dtype: Union[type, str]
+        :return: An empty WolfArray with the given name and header.
+        :rtype: WolfArray
+        """
+        array = WolfArray()
+        array.init_from_header(self.header, dtype=dtype)
+        return array
+
+    def as_results(self,
+                      results: np.ma.MaskedArray,
+                      EPSG: int = 31370,
+                      dtype: Union[type, str] =None,
+                      write_to: Union [Path,str] = None
+                      ) -> WolfArray:
+        """
+        Write the results to a WolfArray file.
+
+        :param output_path: The path to write the results to.
+        :param name: The name of the results file.
+        :param header: The header for the WolfArray.
+        :param dtype: The data type of the WolfArray.
+        """
+        if dtype is None:
+            dtype = results.dtype
+        array = self.create_WolfArray(dtype=dtype)
+        array.array = results
+        array.array.mask = results.mask
+        array.nullvalue = -99999
+        if write_to is not None:
+            assert isinstance(write_to, (str, Path)), \
+                f"The output path must be a string or a Path object, not {type(write_to)}."
+            # if not os.path.exists(write_to):
+            #     os.makedirs(write_to)
+            array.write_all(write_to, EPSG=EPSG)
+        return array
+
+class Input:
+    """
+    This class stores the inputs of the MultiCriteriaAnalysis.
+
+    Also, it ensures that the inputs are valid and consistent
+    to feed the MulticriteriAnalysis tool.
+    """
+    def __init__(self,
+                 name: str,
+                 array: np.ma.MaskedArray| np.ndarray| WolfArray,
+                 condition: Literal[Operator.SUPERIOR.value,
+                                    Operator.INFERIOR.value,
+                                    Operator.EQUAL.value,
+                                    Operator.SUPERIOR_OR_EQUAL.value,
+                                    Operator.INFERIOR_OR_EQUAL.value] = Operator.SUPERIOR_OR_EQUAL.value,
+                threshold: Union[float, int] = 0.0,
+                ) -> None:
+        """
+        Initialize the Input object.
+
+        :param name: The name of the variable.
+        :type name: str
+        :param array: The array (matrix) containing the variable (observations).
+        :type array: np.ma.MaskedArray, np.ndarray, or WolfArray
+        :param condition: The condition (criteria) used to discriminate values inside the variable,
+        defaults to 'superior or equal'.
+        :type condition: str, optional
+        :param threshold: The threshold used to select values inside the variable,
+        defaults to 0.0.
+        :type threshold: Union[float, int], optional
+        """
+        self._name = None
+        self._array = None
+        self._condition = None
+        self._threshold = None
+        self._score = None
+        self.name = name
+        self.array = array
+        self.condition = condition
+        self.threshold = threshold
+
+    @property
+    def name(self) -> str:
+        """
+        Get the name of the input.
+
+        :return: The name of the input.
+        """
+        return self._name
+
+    @name.setter
+    def name(self, value: str) -> None:
+        """
+        Set the name of the input.
+
+        :param value: The name to set.
+        """
+        assert isinstance(value, str), f"The name must be a string, not {type(value)}."
+        self._name = value
+
+    @property
+    def array(self) -> Union[np.ma.MaskedArray, np.ndarray, WolfArray]:
+        """
+        Get the array of the input.
+
+        :return: The array as a numpy masked array.
+        """
+        return self._array
+
+    @array.setter
+    def array(self, value: Union[np.ma.MaskedArray, np.ndarray, WolfArray
+                    ]) -> None:
+        """
+        Set the array of the input.
+
+        :param value: The array to set.
+        """
+        assert isinstance(value, (np.ma.MaskedArray, np.ndarray, WolfArray)), \
+            f"The array must be a numpy masked array, a numpy array or a WolfArray, not {type(value)}."
+        self._array = value
+
+    @property
+    def condition(self) -> Literal[Operator.SUPERIOR.value,
+                                    Operator.INFERIOR.value,
+                                    Operator.EQUAL.value,
+                                    Operator.SUPERIOR_OR_EQUAL.value,
+                                    Operator.INFERIOR_OR_EQUAL.value]:
+        """
+        Get the condition of the input.
+
+        :return: The condition as a string.
+        """
+        return self._condition
+
+    @condition.setter
+    def condition(self, value: Literal[Operator.SUPERIOR.value,
+                                        Operator.INFERIOR.value,
+                                        Operator.EQUAL.value,
+                                        Operator.SUPERIOR_OR_EQUAL.value,
+                                        Operator.INFERIOR_OR_EQUAL.value]) -> None:
+        """
+        Set the condition of the input.
+
+        :param value: The condition to set.
+        """
+        assert isinstance(value, str), f"The condition must be a string, not {type(value)}."
+        self._condition = value
+
+    @property
+    def threshold(self) -> Union[float, int]:
+        """
+        Get the threshold of the input.
+
+        :return: The threshold as a float or an int.
+        """
+        return self._threshold
+
+    @threshold.setter
+    def threshold(self, value: Union[float, int, tuple, list, np.ndarray]) -> None:
+        """
+        Set the threshold of the input.
+
+        :param value: The threshold to set.
+        FIXME make it fair for list as well.
+        """
+        assert isinstance(value, (float, int, tuple, list, np.ndarray)),\
+            f"The threshold must be a float or an int or a tuple, not {type(value)}."
+        if isinstance(value, (tuple, list)):
+            assert len(value) == 2, "The threshold must be a tuple or a list with exactly two elements."
+            assert all(isinstance(item, (float, int)) for item in value), \
+                f"All elements in the threshold tuple or list must be floats or ints.\
+                    not {[type(item) for item in value]}."
+            value = tuple(value)  # Ensure it's a tuple
+        elif isinstance(value, np.ndarray):
+            assert value.ndim == 1, "The threshold must be a 1D numpy"
+            assert len(value) == 2, "The threshold must be a 1D numpy array with exactly two elements."
+            # val1 = value[0]
+            if np.issubdtype(value.dtype, np.integer):
+                value = (int(value[0]), int(value[1]))
+            elif np.issubdtype(value.dtype,np.floating):
+                value = (float(value[0]), float(value[1]))
+            # value = (float(value[0]), float(value[1]))  # Convert to tuple
+        self._threshold = value
+
+    @property
+    def score(self) -> Score:
+        """
+        Get the score of the input.
+
+        :return: The score as a Score object.
+        """
+        if self._score is None:
+            self._score = Score(variable= Variable(variable = self.array),
+                                criteria= Criteria(threshold=self.threshold, condition=self.condition, ))
+        return self._score
+
+
+    @score.setter
+    def score(self, value: Score) -> None:
+        """
+        Set the score of the input.
+
+        :param value: The score to set.
+        """
+        assert isinstance(value, Score), f"The score must be an instance of Score, not {type(value)}."
+        self._score = value
+
+class MulticriteriAnalysis:
+    """
+    This class performs a multi-criteria analysis based
+    on a set of inputs (variables and criteria).
+
+    The inputs are:
+        - name: The name of the variable (observations).
+        - array: The array (matrix) containing the variable (observations).
+        - condition: The condition (criteria) used to distinguish between good and bad variables.
+        Basically, the condition is used to obtain a binary score (0 or 1) for each value in the array.
+        - Threshold: The value(s) used as limits for the condition.
+    """
+    def __init__(self,
+                 inputs: Union[list[Input],tuple[Input],Input],
+                 method:Literal[Operator.SUM.value,
+                                Operator.AVERAGE.value,
+                                Operator.PERCENTAGE.value] = Operator.PERCENTAGE.value,
+                 dtype: Union[type, str] = Format.FLOAT32.value,
+                 mold: WolfArray = None,
+                 write_to: Union[Path, str] = None,
+                 EPSG: int = None
+                 ) -> None:
+        """
+        Initialize the MultiCriteriaAnalysis
+
+        :param inputs: The inputs to the MultiCriteriaAnalysis.
+        :type inputs: Union[list[Input], tuple[Input], Input]
+        :param method: The method used for the MultiCriteriaAnalysis,
+        defaults to 'percentage'.
+        :type method: Literal[constant_value(Constant.SUM),
+                        constant_value(Constant.AVERAGE),
+                        constant_value(Constant.PERCENTAGE)], optional
+        :param dtype: The data type used for the MultiCriteriaAnalysis,
+        defaults to np.float32.
+        :type dtype: Union[type, str], optional
+        :param mold: The mold used for the MultiCriteriaAnalysis,
+        defaults to None.
+        :type mold: WolfArray, optional
+        :param write_to: The path where the results will be written,
+        defaults to None.
+        :type write_to: Union[Path, str], optional
+        :param EPSG: The EPSG code for the coordinate reference system,
+        defaults to None.
+        :type EPSG: int, optional
+        """
+        self._inputs: list[Input] = None
+        self._method: str = None
+        self._dtype: Union[type, str] = None
+        self._mold: WolfArray = None
+        self._path: Union[Path, str] = None
+        self._EPSG: int = None
+        self.inputs = inputs
+        self.method = method
+        self.dtype = dtype
+        self.mold = mold
+        self.path = write_to
+        self.EPSG = EPSG
+
+    # ************************
+    # Preprocessing the inputs
+    # ************************
+
+    @property
+    def inputs(self) -> list[Input]:
+        """
+        Get the inputs of the MultiCriteriaAnalysis.
+
+        :return: The inputs as a list of Input objects.
+        """
+        return self._inputs
+
+    @inputs.setter
+    def inputs(self, value: Union[list[Input],tuple[Input],Input]) -> None:
+        """
+        Set the inputs of the MultiCriteriaAnalysis.
+
+        :param value: The inputs to set.
+        """
+        assert isinstance(value, (list, tuple, Input)), \
+            f"The inputs must be a list or a tuple of Input objects, not {type(value)}."
+        if isinstance(value, (list, tuple)):
+            assert len(value) > 0, "The inputs list must not be empty."
+            assert all(isinstance(item, Input) for item in value), \
+                "All items in the inputs list must be Input objects."
+            if isinstance(value, tuple):
+                value = list(value)
+        elif isinstance(value, Input):
+            value = [value]
+
+        self._inputs = value
+
+    @property
+    def number_of_inputs(self) -> int:
+        """
+        Get the number of inputs.
+
+        :return: The number of inputs.
+        """
+        return len(self.inputs) if self.inputs is not None else 0
+
+    @property
+    def method(self) -> Literal[Operator.SUM.value,
+                                Operator.AVERAGE.value,
+                                Operator.PERCENTAGE.value]:
+        """
+        Get the method used for the MultiCriteriaAnalysis.
+
+        :return: The method as a string.
+        """
+        return self._method
+
+    @method.setter
+    def method(self, value: Literal[Operator.SUM.value,
+                                      Operator.AVERAGE.value,
+                                      Operator.PERCENTAGE.value]) -> None:
+        """
+        Set the method used for the MultiCriteriaAnalysis.
+
+        :param value: The method to set.
+        """
+        assert isinstance(value, str), f"The method must be a string, not {type(value)}."
+        self._method = value
+
+    @property
+    def dtype(self) -> Union[type, str]:
+        """
+        Get the data type used for the MultiCriteriaAnalysis.
+
+        :return: The data type as a string or a numpy dtype.
+        """
+        return self._dtype
+
+    @dtype.setter
+    def dtype(self, value: Union[type, str]) -> None:
+        """
+        Set the data type used for the MultiCriteriaAnalysis.
+
+        :param value: The data type to set.
+        """
+        if value is None:
+            value = Format.FLOAT32.value  # Default data type
+        else:
+            assert isinstance(value, (type, str)),\
+                f"The data type must be a numpy dtype or a string representing a numpy dtype, not {type(value)}."
+            self._dtype = value
+
+    @property
+    def mold(self) -> WolfArray:
+        """
+        Get the mold used for the MultiCriteriaAnalysis.
+
+        :return: The mold as a WolfArray.
+        """
+        return self._mold
+
+    @mold.setter
+    def mold(self, value: WolfArray) -> None:
+        """
+        Set the mold used for the MultiCriteriaAnalysis.
+
+        :param value: The mold to set.
+
+        FIXME: add update mold from inputs.
+        """
+        if value is not None:
+            assert isinstance(value, WolfArray), f"The mold must be a WolfArray, not {type(value)}."
+            for input_ in self.inputs:
+                if value.shape != input_.array.shape:
+                    raise ValueError(f"The mold shape {value.shape} does not match the input shape {input_.array.shape}.")
+        self._mold = value
+
+    @property
+    def path(self) -> Union[Path, str]:
+        """
+        Get the path where the results will be written.
+
+        :return: The path as a string or a Path object.
+        """
+        return self._path
+
+    @path.setter
+    def path(self, value: Union[Path, str]) -> None:
+        """
+        Set the path where the results will be written.
+
+        :param value: The path to set.
+        """
+        if value is None:
+            self._path = None
+        else:
+            assert isinstance(value, (Path, str)),\
+                f"The path must be a string or a Path object, not {type(value)}."
+            self._path = value
+
+    @property
+    def scores(self) -> Scores:
+        """
+        Get the scores for each input.
+
+        :return: The scores as a Scores object.
+        """
+        scores = {}
+        for input_ in self.inputs:
+            scores[input_.name] = input_.score
+
+        return Scores(scores)
+
+    @property
+    def operations(self) -> Operations:
+        """
+        Get the operations to be performed on the scores.
+
+        :return: The Operations object.
+        """
+        return Operations(scores=self.scores, int_type=self.dtype, float_type=self.dtype)
+
+    @property
+    def EPSG(self) -> int:
+        """
+        Get the EPSG code for the results.
+
+        :return: The EPSG code as an integer.
+        """
+        return self._EPSG
+
+    @EPSG.setter
+    def EPSG(self, value: int) -> None:
+        """
+        Set the EPSG code for the results.
+
+        :param value: The EPSG code to set.
+        """
+        if value is None:
+            self._EPSG = Constant.EPSG.value  # Default EPSG code
+        else:
+            assert isinstance(value, int), f"The EPSG code must be an integer, not {type(value)}."
+            self._EPSG = value
+
+    # *************************************
+    # Performing the MultiCriteria Analysis
+    # *************************************
+
+    @property
+    def results(self) -> WolfArray:
+        """
+        Perform the MultiCriteria Analysis on the inputs.
+
+        :return: The results of the analysis as a Results object.
+        """
+        if self.number_of_inputs == 0:
+            raise ValueError("No inputs provided for the MultiCriteria Analysis.")
+        assert self.number_of_inputs >= 2, "At least two inputs are required for the MultiCriteria Analysis."
+        assert self.method in self.operations._available_operations,\
+            f"The method '{self.method}' is not available.\
+                Available methods are: {self.operations._available_operations}"
+        # Create results object
+        results = Results(operations=self.operations, mold=self.mold, method=self.method)
+        return  results.as_WolfArray
+
+    def write_result(self) -> WolfArray:
+        """
+        Write the results to a WolfArray file.
+
+        :return: The results as a WolfArray.
+        """
+        if self.path is None:
+            raise ValueError("No path provided for writing the results.")
+        return self.results.write_all(self.path, self.EPSG)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+