Dynamic information aggregation decision-making methods based on variable precision rough set and grey clustering

Purpose – The purpose of this paper is to construct a dynamic information aggregation decision-making model based on variable precision rough set. Design/methodology/approach – To deal with the dynamic decision-making problems, the grey relational analysis method, grey fixed weight clustering based on the centre triangle whitening weight function and maximum entropy principle is used to establish the dynamic information aggregation decision-making model based on variable precision rough set. The method, to begin with, the grey relational analysis method is used to determine the attributes weights of each stage; taking the proximity of the attribute measurement value and positive and negative desired effect value and the uncertainty of time weight into account, a multi-objective optimisation model based on maximum entropy principle is established to solve the model with Lagrange multiplier method, so that time weights expression are acquired; what is more, the decision-making attribute is obtained by grey fixed weight clustering based on the centre triangle whitening weight function, so that multi-decision-making table with dynamic characteristics is established, and then probabilistic decision rules from multi-criteria decision table are derived by applying variable precision rough set. Finally, a decision-making model validates the feasibility and effectiveness of the model. Findings – The results show that it the proposed model can well aggregate the multi-stage dynamic decision-making information, realise the extraction of decision-making rules. Research limitations/implications – The method exposed in the paper can be used to deal with the decision-making problems with the multi-stage dynamic characteristics, and decision-making attributes contain noise data and the attribute values are interval grey numbers. Originality/value – The paper succeeds in realising both the aggregation of dynamic decision-making information and the extraction of decision-making rules.