ORESTE - Organisation, Rangement Et Synthèse de données rElaTionnEllEs

Overview

d(a_i) ≥ 0. Lower d means closer to the ideal reference. ORESTE uses ordinal criterion importance ranks rather than cardinal weights - easier to elicit from decision-makers. The threshold δ controls indifference/incomparability: large δ → more alternatives declared indifferent.

Output

rank sum, lower is better

Data

Crisp, complete numeric matrix

Weights

Needs a weight source

Size

2+ alternatives, 3-12 criteria works best

Used for

Alternative selection, Supplier evaluation

How it works

1. 1

   Rank alternatives per criterion (ordinal input only).

   $$r_{ij}=\text{rank of }{A}_i\text{ on criterion }j$$

   Pastijn-Leysen 1989, p.118 Eq.(1)

2. 2

   Aggregated position D(a,c_j) using Besson rank averaging with criterion rank r(c_j).

   $$D(A_i,c_j)=(\frac{1}{2}(r_{ij}^R+r(c_j{)}^R){)}^{1/R},R=\text{Minkowski exponent}$$

   Pastijn-Leysen 1989, p.119 Eq.(2)

3. 3

   Global aggregated score S(A_i) = Σ_j R(D(A_i, c_j)) (rank of positions).

   $$S(A_i)=\sum _{j=1}^{n}R(D(A_i,c_j))$$

   Pastijn-Leysen 1989, p.120 Eq.(3)

4. 4

   Ascending ranking by S(A_i) (lower is better).

   $$\text{rank}(A_i)\propto S(A_i)$$

   Pastijn-Leysen 1989, p.121

Look elsewhere when

Assumptions to verify

• Criteria preferences are independent (no synergistic interactions)
• Compensation is acceptable: high score on one criterion can offset low on another
• Decision matrix is complete (no missing values)

Edge cases and pitfalls

• δ tuning: too small → many incomparable pairs; too large → all alternatives indifferent.

Works with

How to cite