Almost optimal manipulation of a pair of alternatives

Jacek Szybowski,Konrad Kułakowski,Sebastian Ernst

DOI: https://doi.org/10.1007/s10898-024-01391-3

2023-04-13

Abstract:The role of an expert in the decision-making process is crucial, as the final recommendation depends on his disposition, clarity of mind, experience, and knowledge of the problem. However, the recommendation also depends on their honesty. But what if the expert is dishonest? Then, the answer on how difficult it is to manipulate in a given case becomes essential. In the presented work, we consider manipulation of a ranking obtained by comparing alternatives in pairs. More specifically, we propose an algorithm for finding an almost optimal way to swap the positions of two selected alternatives. Thanks to this, it is possible to determine how difficult such manipulation is in a given case. Theoretical considerations are illustrated by a practical example.

Artificial Intelligence

What problem does this paper attempt to address?

The paper primarily explores the difficulty of manipulating the ranking of two options in the pairwise comparisons method (PCM) and proposes an algorithm to find the closest pairwise comparison matrix (PCM) approximation that makes the positions of two selected options equal. Specifically, this work achieves this goal through orthogonal projection techniques, thereby determining the difficulty of such manipulation in specific scenarios.

The paper first introduces the historical background of the pairwise comparisons method and its application in multi-criteria decision analysis. It then discusses the potential manipulation issues that this method may face and proposes a mathematical model to calculate the optimal pairwise comparison matrix that makes the weights of two options equal. To achieve this goal, the authors define the "tie space," which is the space composed of all matrices that can produce the same ranking results, and construct an orthogonal basis on this foundation. This basis is then used to calculate the best approximation between the original matrix and the target matrix (i.e., the matrix that makes the weights of two specific options equal).

Additionally, the study introduces an "Ease of Manipulation Index" (EMI) to quantify the extent of changes required to alter the relative positions of two options. This index reflects the average difference in elements between the original matrix and the manipulated matrix.

In summary, this paper aims to develop a method that can evaluate and quantify the difficulty of performing specific types of manipulation in the pairwise comparison process.