All methods

CoCoSo

Available

COmbined COmpromise SOlution. Yazdani et al. (2019)

The idea

Two scores, three readings
A3 GEFCO2.8823A2 Bansard2.7880A7 GETMA2.5191A4 Schneider2.4160A1 Mathez2.0413A6 SAGA1.4431A5 LDI Dimotrans1.2987

ka is the share of the total, kb the size against the weakest provider, kc a λ-mix against the best; k combines the three.

Show as table
Alternativekakbkck
A3 GEFCO0.1804.6401.0002.8823
A2 Bansard0.1754.4730.9732.7880
A7 GETMA0.1653.9510.9152.5191
A4 Schneider0.1633.7210.9062.4160
A1 Mathez0.1313.2450.7242.0413
A6 SAGA0.0972.2250.5401.4431
A5 LDI Dimotrans0.0882.0000.4871.2987

Score k as λ moves

No two lines cross: the order is the same at every λ.

Show as table
Alternativeλ = 0.0λ = 0.1λ = 0.2λ = 0.3λ = 0.4λ = 0.5λ = 0.6λ = 0.7λ = 0.8λ = 0.9λ = 1.0
A1 Mathez2.04572.04522.04452.04372.04272.04132.03952.03692.03282.02572.0097
A2 Bansard2.78952.78932.78912.78882.78852.78802.78732.78642.78502.78252.7770
A3 GEFCO2.88232.88232.88232.88232.88232.88232.88232.88232.88232.88232.8823
A4 Schneider2.43422.43192.42922.42582.42162.41602.40852.39772.38082.35072.2819
A5 LDI Dimotrans1.30841.30721.30581.30401.30171.29871.29461.28881.27971.26351.2265
A6 SAGA1.45371.45241.45081.44881.44641.44311.43881.43241.42261.40511.3651
A7 GETMA2.53032.52892.52722.52512.52252.51912.51442.50782.49742.47902.4376

Data: Logistics providers (Yazdani et al. 2019).

CoCoSo gives each alternative a weighted sum and a sum of weighted powers, reads them three ways (share of the total, size against the weakest, a λ-mix against the best) and combines the three.

When to use it

  • Your data mix units or contain negative values.
  • Results should compare with CRITIC or MEREC then CoCoSo studies.
  • Being last on a criterion should cost more than in SAW.

Algorithm

  1. 1Min-max normalization

    The worst value of each column becomes 0 and the best 1, so negative values work too.

  2. 2Weighted sum and power sum

    P is a sum of powers, not a product; a criterion with weight 0 is left out of P, since 0⁰ = 1 would add 1 to every P.

  3. 3Three readings

    Share of the total, size against the weakest alternative, and a λ-balanced mix against the best (λ = 0.5 by default).

  4. 4Combined score

    Rank by decreasing k.

Used together with

Why these pairs
  • Expert weights, then CoCoSo: Yazdani et al. (2019), the example used here.
  • CRITIC then CoCoSo: Akgül (2021), Turkish banks, the third test example.
  • MEREC then CoCoSo appears in recent Turkish studies (Özel Sönmez & Toktaş 2024).
  • The origin compares CoCoSo with WASPAS and VIKOR (same ranking) and with TOPSIS, COPRAS and EDAS.

Reference example

Yazdani, M.; Zaraté, P.; Zavadskas, E.K.; Turskis, Z. (2019)Matches the printed digits
Details

Yazdani, M.; Zaraté, P.; Zavadskas, E.K.; Turskis, Z. (2019). A combined compromise solution (CoCoSo) method for multi-criteria decision-making problems. Management Decision 57(9), 2501-2519. doi:10.1108/MD-05-2017-0458

Published example, reproduced by the tests: seven logistics providers, five criteria (one cost), expert weights (Tables I, III-VI). S, P, ka, kb and kc match the printed 3 decimals and all seven ranks match; the final k differs by up to 0.0013 (1.2987 against the printed "1.3"), because the paper rounded at another stage. Also tested: the paper's second example (ten green suppliers, Tables X-XI) and Akgül (2021, nine Turkish banks, CRITIC weights), all ranks.

Checked by this site's test suite.

Sources

Show 5 sources
  1. Yazdani et al. (2019). A combined compromise solution (CoCoSo) method for multi-criteria decision-making problems. Management Decision 57(9), 2501-2519. doi:10.1108/MD-05-2017-0458
  2. Yazdani, Zaraté, Zavadskas & Turskis (2019), Management Decision 57(9), 2501-2519 doi:10.1108/MD-05-2017-0458
  3. Akgül (2021), CRITIC-CoCoSo, Ekonomi ve Finansal Araştırmalar Dergisi 3(2), 71-90 dergipark.org.tr/tr/download/article-file/2114456
  4. Özel Sönmez, G., & Toktaş, P. (2024). Supplier selection using the integrated MEREC - CoCoSo methods in a medical device company. Journal of Scientific Reports-A, (056), 116-133. doi:10.59313/jsr-a.1420728
  5. pyDecision 5.1.8, cocoso_method (adds 1 to all S or P when the minimum is 0; not adopted) github.com/Valdecy/pyDecision