All methods

CODAS

Available

COmbinative Distance-based ASsessment. Keshavarz Ghorabaee et al. (2016)

The idea

0.02
0.01: Taxicab on smaller E gaps0.10: Taxicab on larger E gaps
Two distances from the worst point, one pairwise score
S1+1.3914S2+0.3411S3-0.2170S6-0.2481S4-0.5381S5-0.7292

Each supplier is measured from the negative-ideal point ns by Euclidean distance E and Taxicab distance T; H sums the pairwise margins.

Show as table
AlternativeETH
S10.21410.3277+1.3914
S20.14110.2256+0.3411
S30.10060.1901-0.2170
S60.10950.1528-0.2481
S40.08470.1254-0.5381
S50.06660.1210-0.7292

Data: Steelmaking suppliers (Badi et al. 2018).

CODAS scores each alternative by its distance from the worst point, Euclidean first and Taxicab second.

When to use it

  • Strictly positive benefit and cost data.
  • You want pairwise aggregation using two notions of distance.

Algorithm

  1. 1Linear normalization
  2. 2Weighting and negative-ideal point
  3. 3Euclidean and Taxicab distances
  4. 4Relative assessment matrix

    psi as printed in Eq. (8) of the origin, which reproduces every published number in Section 3, although the paper's prose says the opposite (use Taxicab when the Euclidean gap is below tau).

  5. 5Assessment score

    Rank by decreasing H.

Used together with

Why these pairs
  • Expert or AHP weights, then CODAS: Badi, Abdulshahed & Shetwan (2018), the example used here; the origin uses weights from Chakraborty & Zavadskas (2014) and Zavadskas & Turskis (2010).
  • Simulated weight sets for sensitivity, compared with WASPAS, COPRAS, TOPSIS, VIKOR and EDAS (origin, Tables 9-11).
  • Example 2 reuses the office-microclimate data of the ARAS paper (2010).

Reference example

Badi, I.; Abdulshahed, A.M.; Shetwan, A. (2018)Matches the printed digits
Details

Badi, I.; Abdulshahed, A.M.; Shetwan, A. (2018). A case study of supplier selection for a steelmaking company in Libya by using the Combinative Distance-based Assessment (CODAS) model. Decision Making: Applications in Management and Engineering 1(1), 1-12. doi:10.31181/dmame180101b

Published example, reproduced by the tests: six steelmaking suppliers on four criteria, direct cost the only cost, with expert weights (Tables 1-5). E, T and H match to 4 decimals and the rank order over tau (0.01 to 1.00) is identical; Table 3 misprints the third column of ns as 0.0719 for 0.1161, while E, T and H use 0.1161. Also tested: both examples of the origin (Keshavarz Ghorabaee et al. 2016, industrial robots in Tables 1-4, where the headers of Tables 1 and 2 swap tip speed and repeatability, and office microclimate in Tables 5-8), which match E, T, H and rankings. pyDecision uses a different h formula and does not match.

Checked by this site's test suite.

Sources

Show 5 sources
  1. Keshavarz Ghorabaee et al. (2016). A new combinative distance-based assessment (CODAS) method for multi-criteria decision-making. Economic Computation and Economic Cybernetics Studies and Research 50(3), 25-44. https://ideas.repec.org/a/cys/ecocyb/v50y2016i3p25-44.htmlNo DOI found; none is printed in the PDF.
  2. Badi, Abdulshahed & Shetwan (2018), Decision Making: Applications in Management and Engineering 1(1), 1-12 doi:10.31181/dmame180101b
  3. Trung (2022), Applied Engineering Letters 7(2), 54-66 doi:10.18485/aeletters.2022.7.2.2
  4. pymcdm 1.4.0, pymcdm.methods.CODAS gitlab.com/shekhand/mcda
  5. pyDecision 5.1.8, codas_method github.com/Valdecy/pyDecision