An outranking Choquet integral formulation for multi-criteria sorting problems with heterogeneous scales: an extension of FlowSort for interacting criteria

TL;DR

This paper introduces a novel Choquet integral-based method for multi-criteria sorting that handles heterogeneous scales and interacting criteria, extending FlowSort to improve practical decision-making.

Contribution

It extends FlowSort by incorporating a new Choquet integral formulation that manages heterogeneous scales and criterion interactions in sorting problems.

Findings

The proposed method effectively models interacting criteria.

It handles heterogeneous scales without requiring common measurement units.

Numerical examples demonstrate simplicity and importance of considering interactions.

Abstract

In multi-criteria decision aiding, the Choquet integral has been used as an aggregation operator to deal with interacting criteria, having as a requirement a prior assumption of common scale for all the criteria. This restriction on the adopted scale may be considered as a limitation in some practical problems. In order to overcome this limitation, we propose a new Choquet integral formulation, specific for sorting problems, that constructs a common scale from heterogenous scales using the framework of the FlowSort method. We also introduce a new outranking degree based on the Choquet integral preference model, which allows the modeling of interacting criteria in FlowSort. Therefore, the proposed approach can be seen either as an extension of FlowSort for problems with interacting criteria or as a new Choquet integral formulation for multi-criteria sorting problems with heterogeneous scales. Numerical examples attest that the proposed approach is conceptually simple to be implemented and acknowledge the importance of considering interaction between criteria when it exists.