COPRAS
AvailableCOmplex PRoportional ASsessment. Zavadskas et al. (1994)
The idea
Each bar is S+ (benefit share) plus the cost bonus, which grows as the cost share S- shrinks; it ends at Q, then N in percent of the best.
- Benefit share S+
- Cost bonus
Show as table
| Alternative | Benefit share S+ | Cost bonus | Relative significance Q | Utility degree N (%) |
|---|---|---|---|---|
| A4 | 0.0883 | 0.0819 | 0.1702 | 100.00% |
| A7 | 0.0812 | 0.0811 | 0.1623 | 95.38% |
| A6 | 0.1032 | 0.0584 | 0.1615 | 94.92% |
| A5 | 0.0789 | 0.0588 | 0.1377 | 80.90% |
| A2 | 0.0530 | 0.0810 | 0.1340 | 78.72% |
| A3 | 0.0587 | 0.0619 | 0.1206 | 70.87% |
| A1 | 0.0318 | 0.0819 | 0.1137 | 66.82% |
Data: Cream separators (Özcan and Çelik 2021).
COPRAS adds up each alternative's weighted benefit shares, adds a bonus that grows as its cost share shrinks, and shows every alternative as a percentage of the best.
When to use it
- Every value is a positive amount: prices, durations, counts.
- Benefit and cost criteria are mixed.
- Costs should stay in their own term, not flipped first.
Algorithm
1Weighted shares of the column total
Benefit and cost columns alike; nothing is inverted.
2Benefit and cost sums
3Relative significance
With no cost criteria, or cost weights that are all 0, Q equals S+ (SAW with the sum option).
4Utility degree
Rank by decreasing Q; the best alternative has N = 100 %.
Used together with
Why these pairs
- Entropy weights, then COPRAS: Özcan & Çelik (2021), the example used here.
- AHP weights, then COPRAS (Aytaç Adalı & Tuş Işık 2016); the 2007 restatement uses expert weights.
- SWARA and COPRAS co-occur in 154 OpenAlex records, more than SWARA and TOPSIS.
- With at most one cost criterion COPRAS gives exactly SAW with the sum option; they differ from two cost criteria on.
Reference example
Details
Özcan, S.; Çelik, A.K. (2021). A comparison of TOPSIS, grey relational analysis and COPRAS methods for machine selection problem in the food industry of Turkey. International Journal of Production Management and Engineering 9(2), 81-92. doi:10.4995/ijpme.2021.14734
Published example, reproduced by the tests: seven cream separators, seven criteria (four of them cost), Entropy weights (Tables 1, 2, 10, 11). All 49 cells of Table 10 match to 4 decimals and the seven Q values of Table 11 to 0.0001; our Entropy reproduces the weights of Table 2. The printed rank row swaps A2 and A3 (it copies the TOPSIS column); the ranks here follow the printed Q. Also tested: Podvezko (2011, Tables 7-8), and Zavadskas et al. (2007), whose Table 2 prints Q = S+ + S- instead of its own Eq. 5, so only its S+ and S- are used.
Checked by this site's test suite.
Sources
Show 7 sources
- Zavadskas et al. (1994). The new method of multicriteria complex proportional assessment of projects. Technological and Economic Development of Economy 1(3), 131-139. The origin the authors cite themselves; no DOI and no copy online, so it was not opened. The formulas follow their own restatement (Zavadskas, Kaklauskas, Peldschus & Turskis 2007, Eqs. 1-6), checked against Podvezko (2011).
- Zavadskas, Kaklauskas & Šarka (1994), Technological and Economic Development of Economy 1(3), 131-139 (not opened)
- Zavadskas, Kaklauskas, Peldschus & Turskis (2007), Multi-attribute assessment of road design solutions by using the COPRAS method, The Baltic Journal of Road and Bridge Engineering 2(4), 195-203
- Podvezko (2011), The Comparative Analysis of MCDA Methods SAW and COPRAS, Engineering Economics 22(2), 134-146 doi:10.5755/j01.ee.22.2.310
- Özcan & Çelik (2021), International Journal of Production Management and Engineering 9(2), 81-92 doi:10.4995/ijpme.2021.14734
- Keshavarz Ghorabaee, Zavadskas, Olfat & Turskis (2015), Informatica 26(3), 435-451 doi:10.15388/Informatica.2015.57
- pyDecision 5.1.1, copras_method (returns only N / 100) github.com/Valdecy/pyDecision