Topsis-Kirtan-102203600 1.0.0

It Ranks the Models based on MCDM

Project Description

TOPSIS Implementation

This python package is submitted by Kirtan Dwivedi; Roll No.- 102203600

Overview

TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) is a multi-criteria decision-making (MCDM) method. This Python package implements the TOPSIS method, allowing you to rank alternatives based on multiple criteria.

This implementation is specifically designed to work with a CSV input file and provides results with TOPSIS scores and ranks.

Features

• Handles multi-criteria decision-making problems
• Accepts CSV input files
• Outputs a CSV file with TOPSIS scores and ranks
• Validates input data and handles common errors
• Command-line interface for ease of use

Guide Video and Web Service for Topsis.

For a detailed guide, you can watch this Topsis package in python by Kirtan Dwivedi and check this Topsis-Kirtan-102203600 for direct use.

If you have any questions, suggestions, or issues, feel free to comment.

OR

Installation

You can install this package directly from PyPI using pip:

pip install Topsis-Kirtan-102203600

Usage

Command-Line Interface

After installing the package, you can use it from the command line. The basic usage is:

python topsis <InputDataFile> <Weights> <Impacts> <ResultFileName>

Parameters

• InputDataFile: The path to the input XLSX file containing the data.
• Weights: A string of comma-separated numerical weights (e.g., "1,1,1,2").
• Impacts: A string of comma-separated + or - signs indicating whether each criterion is beneficial (+) or non-beneficial (-) (e.g., "+,+,-,+").
• ResultFileName: The path to the output CSV file where results will be saved.

Example

Suppose you have an input file named 102203600_data.xlsx with the following content:

| Alternative | Criterion1 | Criterion2 | Criterion3 | Criterion4 |
|-------------|------------|------------|------------|------------|
| M1          | 250        | 16         | 12         | 5          |
| M2          | 200        | 6          | 8          | 3          |
| M3          | 300        | 16         | 8          | 4          |
| M4          | 275        | 10         | 10         | 4          |

You can run the following command:

topsis 102203600_data.xlsx "1,1,1,2" "+,+,-,+"  102203600_result.csv

This will generate a file 102203600_result.csv containing the input data along with two additional columns for the TOPSIS score and the rank:

| Alternative | Criterion1 | Criterion2 | Criterion3 | Criterion4 | TOPSIS Score | Rank |
|-------------|------------|------------|------------|------------|--------------|------|
| M1          | 250        | 16         | 12         | 5          | 0.556        | 2    |
| M2          | 200        | 6          | 8          | 3          | 0.222        | 4    |
| M3          | 300        | 16         | 8          | 4          | 0.778        | 1    |
| M4          | 275        | 10         | 10         | 4          | 0.444        | 3    |

Error Handling

The program includes several checks to handle common errors:

• Correct number of parameters (inputFileName, Weights, Impacts, resultFileName).
• Show the appropriate message for wrong inputs.
• Handling of “File not Found” exception
• Input file must contain three or more columns.
• From 2nd to last columns must contain numeric values only (Handling of non-numeric values) -Number of weights, number of impacts and number of columns (from 2nd to last columns) must be same.
• Impacts must be either +ve or -ve.
• Impacts and weights must be separated by ‘,’ (comma).

Examples

Example 1: Basic Usage

python topsis data.xlsx "1,1,1,1" "+,+,+,+" result.csv

Example 2: Different Weights and Impacts

python topsis data.xlsx "0.5,1,1.5,2" "+,-,+,-" result.csv

Example 3: Handling Errors

If the number of weights or impacts does not match the number of criteria:

python topsis data.xlsx "1,1,1" "+,+" result.csv

This will result in an error message indicating the mismatch.

Testing

To ensure everything works correctly, you can run some tests. If you haven't already, you can clone the repository and run the tests locally.

# Clone the repository
git clone https://github.com/KirtanDwivedi/Topsis-Kirtan-102203600

# Navigate to the project directory
cd Topsis_Kirtan_102203600

# Install the required dependencies
pip install -r requirements.txt

# Run the tests (assuming you've created test scripts)
python -m unittest discover tests

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

Inspired by the original TOPSIS method developed by C.L. Hwang and K. Yoon.