const linearAlgebra = require('linear-algebra')();
const Matrix = linearAlgebra.Matrix; `
	
const topsis = require('topsis');

This is the first npm library in NodeJS to implement TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution). This library is fully open-source and free to use. TOPSIS is an algorithm developed in 1981 by Hwang and Yoon. It is a technique commonly utilised in the field of operations research, managerial accounting, decision science and prescriptive analytics for multi-attribute decision-making problems (MADC). Its most common use case is to select the optimal choice from an array of alternatives with conflicting criteria. (For instance, to select a celphone to buy that have the highest quality camera at the lowest price). The central idea behind the algorithm is that the most desirable solution is the one that has the highest geometric distance (euclidean distance) to the anti-ideal solution and the lowest euclidean distance to the ideal solution.

npm i topsis

let m = new Matrix([
                    [2, 5, 5], 
                    [60, 26, 4], 
                    [20, 20, 4],  
                    [500, 2, 4], 
                    [50, 23, 3], 
                    [25, 10, 1]
                    ]); 
// m argument is the alternative matrix. Each row is an alternative and each column is a criterion.
                    
let w = [0.27,0.33,0.40];  // This argument indicates the weights of each criteria.
let ia = ['min', 'min', 'max']; // This argument indicates if a criterion is beneficial or not.


topsis.getBest(m, w, ia);

Returns:

[2, 5, 5]

https://runkit.com/patelotech/runkit-npm-topsis

NOTE: Make sure to read through bibliography below of TOPSIS to understand how the algorithm works.

1. Computing normalised alternative matrix. (Alternative matrix divided by calculated norm of each column).
2. Computing weighted normalised alternative matrix. (Alternative matrix multiplied by eigenvector).
3. Calculating the ideal solution. (An hypothetical alternative that maximises benefits and minimises costs the most).
4. Calculating the anti-ideal solution. (An hypothetical alternative that minimises benefits and maximises costs the most).
5. For each alternative in the weighted normalised alternative matrix, calculate the euclidean distance to the ideal and anti-ideal solution.
6. Calculate a performance score for each alternative. (Distance to the anti-ideal solution divided by the sum of the distance of the ideal and anti-ideal solution).
7. Rank in descending order performance scores for each alternative.
8. Return the top-one.

topsis.createRandom(); // Creates random arguments for a topsis.getBest function.

Returns:

{ 
  m : 
    { data:
     [ [ 478, 6, 184, 1000 ],
       [ 805, 34, 703, 770 ],
       [ 216, 210, 294, 225 ],
       [ 857, 253, 396, 63 ],
       [ 244, 652, 886, 736 ],
       [ 378, 824, 948, 769 ],
       [ 298, 76, 573, 604 ] ],
      rows: 7,
      cols: 4 
    },
  w: [ 0.52, 0.26, 0.04, 0.18 ],
  ia: [ 'max', 'min', 'max', 'max' ]
}

• TOPSIS.
• Support for Linear-Algebra matrices.
• Support for MathJS matrices.
• (FTOPSIS) Fuzzy TOPSIS
• (IFTOPSIS) Intuisionistic Fuzzy TOPSIS
• (NTOPSIS) Neutrosophic TOPSIS

• linear-algebra

MIT

https://github.com/patelotech/topsis

• AIRBNB AIRBNB JS CODE STYLE

• Eslint v-4.19.1 // AIRBNB Configuration

• Error check: npm run check
• Error fix: npm run lint or npm run lint -- --fix

• Hwang, C. L., & Yoon, K. (1981). Methods for multiple attribute decision making. In Multiple attribute decision making (pp. 58-191).Springer, Berlin, Heidelberg. cd do cd do