Novel method of solving attribute weights in multi-attribute decision making

Aiming at the problem of how to solve the attribute weights in multi-attribute decision making, this paper presents a new method of solving attribute weights based on the optimal membership and relative entropy. Firstly, the definitions of optimal membership and relative entropy are given. Secondly, by solving a series of linear programming models whose goals are to maximize the optimal membership, the preference weight vectors which are partial for each alternative are obtained, and then the optimal weight vector is got by establishing a relative entropy weight model based on these preference weight vectors. Finally, an example of solving attribute weights is given to demonstrate the feasibility and rationality of this new method.