AHP-CALC

A TypeScript module implementing the Analytical Hierarchy Process (AHP) for decision support systems.
It supports pairwise comparison matrices, calculates priority weights, and evaluates consistency.
Ideal for multi-criteria decision analysis (MCDA).

⚠️ Note: This library does not support sub-criteria (yet).

---

📋 Requirements

This library requires:

• A pairwise comparison matrix for the criteria.
• A pairwise comparison matrix for the alternatives with respect to each criterion.

---

📦 Installation

npm install ahp-calc

Documentation

Documentation

Main Feature

• Support for Pairwise Comparison Matrices: The library allows you to create and process pairwise comparison matrices for both criteria and alternatives.

• Matrix Normalization: Supports normalization of both criteria and alternatives matrices to calculate relative priorities.

• Weight Calculation (bobot priority): It calculates local priority weights (eigenvectors) for criteria and alternatives using the normalized matrices, and it can compute the global weights across all levels.

• Consistency Evaluation: Includes functionality for consistency checking using λmax (Lamda Max), Consistency Index (CI), and Consistency Ratio (CR) to evaluate whether the matrix comparisons are consistent.

• Converts Nested String Matrices to Numeric Matrices: The library supports converting string-based matrices (nested) into numeric matrices, making it easier to process data in AHP.

• Priority Weights Calculation for Alternatives: For multi-alternative problems, the library supports calculating priority weights for alternatives against each criterion.

• Consistency Check for Alternatives: The library can evaluate the consistency of matrices for alternatives as well, using the same λmax, CI, and CR methods.

No Support for Sub-criteria: ⚠️ Currently, the library does not support sub-criteria for further hierarchical decision-making.

Example Usage

you can start directly using function to calculate the pairwaise

for example to calculate Criteria Pairwise Comparison Matrices:

import { calculcateCritMatrix } from "ahp-calc";

const critMatriks = [
  ["1", "3", "5", "7", "7"],
  ["1/3", "1", "3", "5", "5"],
  ["1/5", "1/3", "1", "3", "3"],
  ["1/7", "1/5", "1/3", "1", "2"],
  ["1/7", "1/5", "1/3", "1/2", "1"],
];

const {
  normalizedMatrix,
  CI,
  CR,
  RI,
  originalMatrix,
  konsistensi,
  lamdaMax,
  sumCrit,
  n,
  weightsCriteria,
} = calculcateCritMatrix(critMatriks);
console.log({
  sumCrit,
  normalizedMatrix,
  CI,
  CR,
  RI,
  originalMatrix,
  lamdaMax,
  konsistensi,
  n,
  weightsCriteria,
});

for example to calculate alternatif Pairwise Comparison Matrices and if you just wan to take normalize you can do like this:

import { calculateAltMatrix } from "ahp-calc";
const altMatrix[] = [
  [
    ["1", "3", "2", "1/3"],
    ["1/3", "1", "1/3", "1/5"],
    ["1/2", "3", "1", "1/3"],
    ["3", "5", "3", "1"],
  ],
  [
    ["1", "3", "5", "2"],
    ["1/3", "1", "3", "1/3"],
    ["1/5", "1/3", "1", "1/5"],
    ["1/2", "3", "5", "1"],
  ],
  [
    ["1", "1/3", "1/5", "1/5"],
    ["3", "1", "1/3", "1/3"],
    ["5", "3", "1", "2"],
    ["5", "3", "1/2", "1"],
  ],
  [
    ["1", "1/3", "2", "1/3"],
    ["3", "1", "3", "2"],
    ["1/2", "1/3", "1", "1/3"],
    ["3", "1/2", "3", "1"],
  ],
];

const { normalized } = calculateAltMatrix(altMatrix);

console.log(normalized);

Or you can start using the library by first creating an instance of the AHPCrit or AHPAlt class depending on whether you're working with criteria or alternatives.

import { AHPCrit } from "ahp-calc";

const critMatrix = [
  [
    [1.0, 3.0, 0.2],
    [0.3333, 1.0, 0.1429],
    [5.0, 7.0, 1.0],
  ],
];

const weights = AHPCrit.calculateCriteriaWeight(critMatrix);
console.log(weights);

To count each column in all matriks for alternatives:

import { AHPAlt } from "ahp-calc";

const originalMatrix = [
  [
    [1.0, 3.0, 0.5],
    [0.3333, 1.0, 0.1429],
    [2.0, 7.0, 1.0],
  ],
  [
    [1.0, 3.0, 0.5],
    [0.3333, 1.0, 0.1429],
    [2.0, 7.0, 1.0],
  ],
  [
    [1.0, 3.0, 0.5],
    [0.3333, 1.0, 0.1429],
    [2.0, 7.0, 1.0],
  ],
];

const sumAlt = AHPAlt.countTotalAlterEachColumn(originalMatrix);
console.log(sumAlt);

Ranks alternatif

const { weightsCriteria } = calculcateCritMatrix(critMatriks);
const { weightAlt } = calculateAltMatrix(altMatrix);
const res = calculateCompositeWeights(weightAlt, weightsCriteria);
console.log(res);

📊 Example AHP Calculation Output

This is an example output from running a criteria comparison using the AHP (Analytic Hierarchy Process) method:

🔢 Input Matrix (Pairwise Comparisons)

0	1	2	3	4
1	3	5	7	7
0.333	1	3	5	5
0.2	0.333	1	3	3
0.143	0.2	0.333	1	2
0.143	0.2	0.333	0.5	1

📉 Normalized Matrix

0	1	2	3	4
0.5498	0.6338	0.5173	0.4242	0.3889
0.1831	0.2113	0.3104	0.3030	0.2778
0.1100	0.0704	0.1035	0.1818	0.1667
0.0786	0.0423	0.0345	0.0606	0.1111
0.0786	0.0423	0.0345	0.0303	0.0556

⚖️ Calculated Weights

These represent the relative importance of each criterion:

Weight
0.4535
0.2826
0.1486
0.0725
0.0429

📈 Summary

• Column Totals (for normalization): [1.819, 4.733, 9.666, 16.5, 18]
• λ max: 5.277
• Consistency Index (CI): 0.069
• Consistency Ratio (CR): 0.062
• Random Index (RI): 1.12
• ✅ Consistency Check: Matrix is consistent (CR ≤ 0.1)
• Number of criteria: 5

---

AHP Alternative Calculation Results

Matrix Order (n):

[ 4, 4, 4, 4, 4 ]

Original Matrices:

• Matrix [0]:

(index)	0	1	2	3
0	1	3	2	0.333
1	0.333	1	0.333	0.2
2	0.5	3	1	0.333
3	3	5	3	1

• Matrix [1]:

(index)	0	1	2	3
0	1	3	5	2
1	0.333	1	3	0.333
2	0.2	0.333	1	0.2
3	0.5	3	5	1

• Matrix [2]:

(index)	0	1	2	3
0	1	0.333	0.2	0.2
1	3	1	0.333	0.333
2	5	3	1	2
3	5	3	0.5	1

• Matrix [3]:

(index)	0	1	2	3
0	1	0.333	2	0.333
1	3	1	3	2
2	0.5	0.333	1	0.333
3	3	0.5	3	1

• Matrix [4]:

(index)	0	1	2	3
0	1	3	3	3
1	0.333	1	2	2
2	0.333	0.5	1	2
3	0.333	0.5	0.5	1

Normalized Matrices:

• Matrix [0]:

(index)	0	1	2	3
0	0.207	0.25	0.316	0.178
1	0.069	0.083	0.053	0.107
2	0.103	0.25	0.158	0.178
3	0.621	0.417	0.474	0.536

• Matrix [1]:

(index)	0	1	2	3
0	0.492	0.409	0.357	0.566
1	0.164	0.136	0.214	0.094
2	0.098	0.045	0.071	0.057
3	0.246	0.409	0.357	0.283

• Matrix [2]:

(index)	0	1	2	3
0	0.071	0.045	0.098	0.057
1	0.214	0.136	0.164	0.094
2	0.357	0.409	0.492	0.566
3	0.357	0.409	0.246	0.283

• Matrix [3]:

(index)	0	1	2	3
0	0.133	0.154	0.222	0.091
1	0.4	0.462	0.333	0.546
2	0.067	0.154	0.111	0.091
3	0.4	0.231	0.333	0.273

• Matrix [4]:

(index)	0	1	2	3
0	0.5	0.6	0.462	0.375
1	0.167	0.2	0.308	0.25
2	0.167	0.1	0.154	0.25
3	0.167	0.1	0.077	0.125

Column Totals for Each Alternative:

• Column Total [0]: [ '4.833', '12.000', '6.333', '1.866' ]
• Column Total [1]: [ '2.033', '7.333', '14.000', '3.533' ]
• Column Total [2]: [ '14.000', '7.333', '2.033', '3.533' ]
• Column Total [3]: [ '7.500', '2.166', '9.000', '3.666' ]
• Column Total [4]: [ '1.999', '5.000', '6.500', '8.000' ]

Weights from Alternative Matrices:

• Weight [0]: [ '0.253', '0.075', '0.193', '0.479' ]
• Weight [1]: [ '0.409', '0.173', '0.064', '0.353' ]
• Weight [2]: [ '0.064', '0.173', '0.409', '0.353' ]
• Weight [3]: [ '0.164', '0.403', '0.097', '0.336' ]
• Weight [4]: [ '0.465', '0.248', '0.178', '0.109' ]

Lambda Max:

[ '4.132', '4.148', '4.148', '4.152', '4.140' ]

Consistency Index (CI):

[ '0.044', '0.049', '0.049', '0.051', '0.047' ]

Random Index (RI): 0.9

Consistency Ratio (CR):

• CR [0]: 0.049 → Consistent? ✅ Yes
• CR [1]: 0.054 → Consistent? ✅ Yes
• CR [2]: 0.054 → Consistent? ✅ Yes
• CR [3]: 0.057 → Consistent? ✅ Yes
• CR [4]: 0.052 → Consistent? ✅ Yes

Overall Consistency Status: ✅ All Consistent

CONTRIBUTING

please read our to start contribute Contributing Guidelines.