dssystem 0.1.5

The basic concept SAW method is to find the sum of the weighted performance rating for each alternative on all attributes. SAW method requires a process of normalizing the decision matrix (X) to a scale that can be compared with all the ratings of existing alternatives.

About this package

The basic concept SAW method is to find the sum of the weighted performance rating for each alternative on all attributes. SAW method requires a process of normalizing the decision matrix (X) to a scale that can be compared with all the ratings of existing alternatives.

Depedencies

• Python >= 3
• numpy
• pandas

Function

init (data,weights, non_beneficial = None)

Initializer provides 2 required parameters and 1 optional parameter, data is the dataset with DataFrame format, weights is the values that provided to get a optimal alternative, non_beneficial is the column that has type "cost", not benefit.

getDecisionMatrix

Function that return formatted matrix from dataset

normalize

Function that return normalized matrix from decision matrix

createDecision

Function that return the list of alternative's score

getChosenOneByIndex

Function that return the chosen one from the alternatives

Example case

We will simulate "Selection of land - Water resources management" which has 4 criteria(s) : Rainfall, Drainage, Usage of land, Tophography. Usage of land is the only one non beneficial criteria.

• In this case we'll use the weight values like this :

[0.25, 0.25, 0.25, 0.25]

• The example dataset that we had

[[25, 67, 7, 20],
[21, 78, 6, 24],
[19, 53, 5, 33], [22, 25, 2, 31]]

Example of code

from dssystem.method import SimpleAdditiveWeighted
import numpy as np
import pandas as pd
dataset = pd.DataFrame({"Rainfall" : [25, 21, 19, 22],
						"Drainage" : [67, 78, 53, 25],
						"Usage of land" : [7, 6, 5, 2],
						"Tophography" : [20, 24, 33, 31]}, 
						index=["L1","L2","L3","L4"])
						
method = SimpleAdditiveWeighted(dataset, [.25, .25 , .25, .25], ["Usage of land"])
print(method.getChosenOneByIndex()) #to get chosen alternative name
print(method.getDecisionMatrix()) #to get decision matrix
print(method.normalize()) #to get normalized decision matrix
print(method.createDecision()) #to get list of alternative's score

Example of output

L4 #L4 is the chosen one

#output of the decision matrix
[[25 67  7 20]
 [21 78  6 24]
 [19 53  5 33]
 [22 25  2 31]]

#output of normalized decision matrix
[[1.   0.86 0.29 0.61]
 [0.84 1.   0.33 0.73]
 [0.76 0.68 0.4  1.  ]
 [0.88 0.32 1.   0.94]]

#output of list of alternative's score
[0.69 0.72 0.71 0.79]