A novel algorithm for fusing preference orderings by rank-ordered agents

Yager proposed an algorithm to combine multi-agent preference orderings of several alternatives into a single consensus fused ordering, when the agents’ importance is expressed through a rank-ordering and not a set of weights. This algorithm is simple and automatable but has some limitations which reduce its range of application, e.g., (i) preference orderings should not include incomparabilities between alternative and/or omissions of some of them, and (ii) the fused ordering may sometimes not reflect the majority of the multi-agent preference orderings. The aim of this article is to present an enhanced version of the Yager’s algorithm, which overcomes the above limitations. Some practical examples support the description of the new algorithm.