Unequal-interval multivariate MGM (1, n) model for material properties analysis has defects including low precision and adaptability. Utilizing new-information theory and grey system modeling method, this paper proposes an unequal-interval multivariate new-information optimization MGM (1, n) model for material properties analysis. The m-th component of original data sequence is regarded as the initial condition of grey differential equation, initial correction as the design variable, and the mean relative error as objective function. Background value of unequal new-information MGM (1, n) model is studied. Based on exponential, integral characteristic of grey model, and the new-information theory, inhomogeneous exponential function was utilized to fit one-accumulation generating sequence, thus proposing reconstruction method for background value of new-information MGM (1, n) mode. The construction equation of background value is also treated here. With high precision and adaptability, the model is suitable for both equal-interval and unequal-interval modeling. Examples have verified the practicality and reliability of the model. © 2016, UK Simulation Society. All rights reserved.
