SIR - Superiority and Inferiority Ranking

Overview

This manifest selects the valid SIR-SAW complete-ranking variant: superiority and inferiority indexes are aggregated by criterion weights, and alternatives are ranked by n-flow descending. Omit preference_functions only when the Type-I usual criterion is intended.

Output

preference flow, higher is better

Data

Crisp, complete numeric matrix

Weights

Needs a weight source

Size

2+ alternatives, 3-12 criteria works best

Used for

Complex multi-stakeholder decisions

Look elsewhere when

Assumptions to verify

• Decision-maker can specify preference (p), indifference (q), and veto (v) thresholds
• Non-compensatory preference structure

Edge cases and pitfalls

• SIR is a family, not one universal scalar ranking. This implementation is explicitly SAW aggregation plus net-flow complete ranking; Xu's partial intersection ranking and other aggregation procedures are not silently substituted.

• Threshold symbols follow the original generalized-criterion parameterization used here: Level reaches strict preference after q+p, while Linear reaches it after s+r. The p and r inputs are widths, not absolute upper thresholds.

• Missing or invalid thresholds must not collapse V-shape, Level, Linear, or Gaussian into the usual preference function. The kernel rejects such inputs.

Works with

How to cite