Scribd
Upload
8 views

AHP Python Code

The document describes a process for performing a pairwise comparison and calculating weights for various criteria using a matrix. It includes steps for normalizing the matrix, calculating the consistency index and ratio, and determining if the comparisons are consistent. Additionally, it outlines a method for reading raster data, reclassifying values, and applying weights to generate a weighted overlay, which is then saved as a new raster file.

Uploaded by

Aparajita Ghosh
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
8 views

AHP Python Code

The document describes a process for performing a pairwise comparison and calculating weights for various criteria using a matrix. It includes steps for normalizing the matrix, calculating the consistency index and ratio, and determining if the comparisons are consistent. Additionally, it outlines a method for reading raster data, reclassifying values, and applying weights to generate a weighted overlay, which is then saved as a new raster file.

Uploaded by

Aparajita Ghosh
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

import numpy as np

# Define the pairwise comparison matrix

pairwise_matrix = np.array([

[1, 3, 3, 2, 5, 5, 7],

[1/3, 1, 3, 2, 3, 3, 5],

[1/3, 1/3, 1, 1/2, 2, 2, 3],

[1/2, 1/2, 2, 1, 5, 5, 7],

[1/5, 1/3, 1/3, 1/5, 1, 1/2, 5],

[1/5, 1/3, 1/2, 1/5, 2, 1, 5],

[1/7, 1/5, 1/3, 1/7, 1/5, 1/5, 1]

])

# Step 1: Normalize the pairwise comparison matrix

column_sums = pairwise_matrix.sum(axis=0)

normalized_matrix = pairwise_matrix / column_sums

# Step 2: Calculate the weights (priority vector)

weights = normalized_matrix.mean(axis=1)

# Step 3: Calculate λ_max

weighted_sum = pairwise_matrix @ weights

lambda_max = np.mean(weighted_sum / weights)

# Step 4: Calculate the Consistency Index (CI)

n = pairwise_matrix.shape[0] # Number of criteria

CI = (lambda_max - n) / (n - 1)

# Step 5: Calculate the Consistency Ratio (CR)


RI_dict = {1: 0, 2: 0, 3: 0.58, 4: 0.9, 5: 1.12, 6: 1.24, 7: 1.32, 8: 1.41, 9: 1.45}

RI = RI_dict[n] # Random consistency index for 7 criteria

CR = CI / RI if RI else 0

# Display results

print("Pairwise Comparison Matrix:")

print(pairwise_matrix)

print("\nNormalized Matrix:")

print(normalized_matrix)

print("\nWeights (Priority Vector):")

for i, weight in enumerate(weights):

print(f"Criterion {i+1}: {weight:.4f}")

print(f"\nλ_max: {lambda_max:.4f}")

print(f"Consistency Index (CI): {CI:.4f}")

print(f"Consistency Ratio (CR): {CR:.4f}")

if CR < 0.1:

print("The pairwise matrix is consistent.")

else:

print("The pairwise matrix is not consistent. Please revise the comparisons.")


import rasterio

import numpy as np

# Define the file paths for each raster layer (update these with your file paths)

rasters = {

'PISR': 'path_to_pISR.tif',

'MAAT': 'path_to_MAAT.tif',

'LST': 'path_to_LST.tif',

'TWI': 'path_to_TWI.tif',

'LULC': 'path_to_LULC.tif',

'Slope': 'path_to_Slope.tif',

'Aspect': 'path_to_Aspect.tif'

# Define the priority weights for each criterion

weights = {

'PISR': 0.27,

'MAAT': 0.11,

'LST': 0.37,

'TWI': 0.08,

'LULC': 0.026,

'Slope': 0.055,

'Aspect': 0.068

# Function to read a raster file and return its values as a numpy array

def read_raster(raster_path):

with rasterio.open(raster_path) as src:


return src.read(1), src.transform, src.crs

# Function to reclassify raster values from 1-4 scale to 1-5 scale

def reclassify_to_5_scale(raster_array):

# Assume that 1 is the strongest correlation and 4 is the weakest

# Mapping the raster values (1-4) to the new scale (1-5)

reclassified = np.interp(raster_array, [1, 4], [1, 5])

return reclassified

# Initialize an empty list to store the weighted layers

weighted_layers = []

# Loop through each raster, reclassify and weight

for criterion, raster_path in rasters.items():

# Read the raster

raster_array, transform, crs = read_raster(raster_path)

# Reclassify to scale of 1-5

reclassified_raster = reclassify_to_5_scale(raster_array)

# Weight the raster

weighted_raster = reclassified_raster * weights[criterion]

# Append the weighted raster to the list

weighted_layers.append(weighted_raster)

# Stack all the weighted layers and sum them up

total_weighted_overlay = np.sum(weighted_layers, axis=0)


# Normalize the output if necessary (optional, depending on how you want the final scale)

# For example, if you want the final result to stay between 1 and 5, use np.clip()

total_weighted_overlay = np.clip(total_weighted_overlay, 1, 5)

# Save the result as a new raster

output_path = 'path_to_output_weighted_overlay.tif'

with rasterio.open(rasters['PISR']) as src:

# Get metadata from the first raster

metadata = src.meta

metadata.update(dtype=rasterio.float32, count=1)

# Write the output raster

with rasterio.open(output_path, 'w', **metadata) as dst:

dst.write(total_weighted_overlay.astype(rasterio.float32), 1)

print(f"Weighted overlay has been saved to {output_path}")

You might also like