Fusion of multi-agent preference orderings in an ordinal semi-democratic decision-making framework

• The problem of aggregating multi-agent preference orderings into a single fused ordering is analyzed.
• The agents’ importance is expressed through a rank-ordering and not a set of weights.
• An enhanced version of an algorithm proposed by Yager is presented.
• The new algorithm better reflects the multi-agent preference orderings and is more versatile.

This paper focuses on the problem of combining multi-agent preference orderings of different alternatives into a single fused ordering, when the agents’ importance is expressed through a rank-ordering and not a set of weights. An enhanced version of the algorithm proposed by Yager (2001) is presented. The main advantages of the new algorithm are that: (i) it better reflects the multi-agent preference orderings and (ii) it is more versatile, since it admits preference orderings with omitted or incomparable alternatives. The description of the new algorithm is supported by a realistic example.