SWARA
AvailableStep-wise Weight Assessment Ratio Analysis. Keršulienė et al. (2010)
The idea
Starting from q₁ = 1 on the most important criterion, each next criterion divides the previous q by k_j = 1 + s_j; the bars on the right normalize q into weights w summing to 1.
Show as table
| Criterion | Step s | Coeff. k | Chain q | Weight w |
|---|---|---|---|---|
| Expedition of dispute resolution | — | 1.0000 | 1.0000 | 0.2174 |
| Price of dispute resolution | 0.1500 | 1.1500 | 0.8696 | 0.1890 |
| Possibility to appeal | 0.0400 | 1.0400 | 0.8361 | 0.1818 |
| Assurance of confidentiality | 0.2900 | 1.2900 | 0.6482 | 0.1409 |
| Authority of person solving the dispute | 0.0200 | 1.0200 | 0.6354 | 0.1381 |
| Legal advice | 0.0400 | 1.0400 | 0.6110 | 0.1328 |
Data: Dispute resolution (Keršulienė et al. 2010).
SWARA sorts the criteria by importance, then asks how much less important each one is than the one above (0.15 = 15% less).
When to use it
- Experts agree on a ranking and on the step sizes.
- Ranking plus "a bit or much less" is easier than pairwise.
- No consistency check needed: SWARA is consistent by construction.
Algorithm
1Sort and state the steps
Criteria sorted from most to least important, and s_j says how much more important criterion j-1 is than criterion j, as a ratio (0.15 = 15%).
2Recalculated weights
The origin paper calls this w and the final weight q; the letters are swapped here to match the other methods.
3Final weights
Weights come out in the sorted order and are mapped back to the input order.
Used together with
Why these pairs
- SWARA with Entropy as combined subjective and objective weights, groundwater vulnerability: Torkashvand et al. (2020), as listed in Keshavarz-Ghorabaee et al. (2021).
- SWARA then WASPAS is the largest SWARA pairing in OpenAlex (224 co-mentions, more than SWARA and TOPSIS); a Lithuanian-school pattern.
- The origin paper derives SWARA weights for six dispute-resolution criteria (Table 1); the workbench pairs them with an illustrative 4 x 6 rating matrix to demonstrate ranking.
Reference example
Details
Keršulienė, V.; Zavadskas, E.K.; Turskis, Z. (2010). Selection of rational dispute resolution method by applying new step-wise weight assessment ratio analysis (SWARA). Journal of Business Economics and Management 11(2), 243-258. doi:10.3846/jbem.2010.12
Published example (Table 1): six criteria for choosing a dispute resolution method. k, q and w match the printed 2 decimals (largest gap 0.0028 on w); the 4 x 6 alternative matrix in the workbench is illustrative.
Checked by this site's test suite.
Sources
Show 4 sources
- Keršulienė et al. (2010). Selection of rational dispute resolution method by applying new step-wise weight assessment ratio analysis (SWARA). Journal of Business Economics and Management 11(2), 243-258. doi:10.3846/jbem.2010.12
- Keršulienė, Zavadskas & Turskis (2010), Journal of Business Economics and Management 11(2), 243-258 doi:10.3846/jbem.2010.12
- Keshavarz-Ghorabaee et al. (2021), Symmetry 13(4), 525 doi:10.3390/sym13040525
- Torkashvand, Neshat, Javadi & Yousefi (2021), Environmental Science and Pollution Research 28, 46704-46724 doi:10.1007/s11356-020-11406-7