A Robust Minimum-Cost Consensus Model With Uncertain Aggregation Weights Based on Data-Driven Method

Abstract

Due to the existence of cognitive bias, it is difficult to accurately obtain the decision-makers’ weights in the aggregating process, which leads to great fluctuations in the optimal decision-making. Therefore, the purpose of this article is to establish a data-driven robust cost consensus model induced by uncertain weights to deal with the adverse effects of uncertainty in the aggregating process. First, the probability density function of the uncertain weight in the aggregation operator is obtained by the kernel density estimation method, and the confidence level is used to control the perturbation range of the uncertainty. Second, a new uncertainty set based on the data-driven robust optimization method is established, which is called the flexible uncertainty set, and two classes of data-driven robust consensus models under uncertain environments are proposed. In the application of power emergency management, by using the data-driven method to traverse the historical data of weights, the new model can effectively deal with the uncertainty caused by the deviation and improve the quality of the aggregation operator, which proves the applicability of the proposed model.