TOPSIS

Technique for Order of Preference by Similarity to Ideal Solution

Citation

If you use this package, please refer cite.

Installation

INSTALLING VIA PIP

inbuilt Python package management system, pip.

You can can install, update, or delete the topsis2.

install

pip install topsis2

update

pip install --upgrade topsis2

uninstall

pip uninstall topsis2

Usage

Using this TOPSIS implementation is straightforward as importing it and writing only two lines. The important thing is the decision matrix in the type of pandas data frame.

The decision matrix would be some data frame as the following example.

After building your decision matrix, you need to define the criteria types (benefit or cost). To have the type, you can define a list as the impact. For instance, we assume that the first two criteria are benefit criteria and the last is the cost.

impact = ['+', '+', '-']

The ultimate step is assigning the weight array.

weight = np.array([0.1, 0.7, 0.2])

After having the three parameters, the model produces the ranking matrix.

import numpy as np

import pandas as pd

from topsis import topsis



array = np.random.randint(10, size=(5, 3))



decision_matrix = pd.DataFrame(array, columns=[

                               'criterion_' + str(i) for i in range(1, 4)],

                               index=['option_'+str(i) for i in range(1, 6)])



impact = ['+', '+', '-']

weight = np.array([0.1, 0.7, 0.2])



tp = topsis(decision_matrix=decision_matrix,

            weight=weight, impact=impact)

tp.rank()

Requirements

This package takes advantage of the following libraries, which had already imported to the TOPSIS package:

• scipy

• numpy

• pandas

Keywords

TOPSIS, MCDM, MADM

Version

0.0.2

Updated

2022-05-19

Date-released

2022-05-18

More about TOPSIS

To read more about the TOPSIS method and the logic behind the model, please refer here.