Ranking
OWA: Ordered Weighted Averaging
Yager, R. R. · 1988
Strengths
- •Generalises WLC and Boolean operations under a single parameterised framework
- •ORness and trade-off provide transparent diagnostics of the aggregation strategy
- •Order weights allow explicit modelling of optimistic (OR) or pessimistic (AND) decision attitudes
- •Widely implemented in GIS environments (IDRISI, ArcGIS, CommonGIS)
Limitations
- •Rank reversal: adding/removing alternatives changes the reordered weight assignment
- •Assumes criteria are preference-independent (additivity assumption)
- •Order weight determination can be non-trivial; maximum entropy or RIM quantifier methods available
- •Full compensability when ORness > 0: very poor performance on one criterion can be offset
Method assistant
Grounded explanations: it explains the method, it does not compute.
Assumptions to verify
- •Criteria preferences are independent (additivity/linearity assumption holds)
- •Decision maker can articulate risk attitude via order weights (or accept uniform λ = WLC)
- •Criterion values are positive and non-zero for all benefit criteria (required by linear-max value function)
- •Value function correctly represents criterion preference structure before OWA aggregation
When not to use
- •Criteria are strongly interdependent → consider ANP or DEMATEL-weighted SAW
- •Strict non-compensatory preferences → use ELECTRE or Boolean conjunctive/disjunctive screening
- •Decision maker cannot specify risk attitude or order weights → use SAW as simpler baseline
Commonly paired with
- •AHP + OWA (high)
- •ENTROPY + OWA (medium)
- •CRITIC + OWA (medium)
How to cite
Yager, R. R. (1988). On ordered weighted averaging aggregation operators in multicriteria decision making. IEEE Transactions on Systems, Man, and Cybernetics. https://doi.org/10.1109/21.87068