Grey forecast method and artificial neural network have their own advantages and disadvantages in building deformation prediction. In order to improve the prediction accuracy,in this paper,combined with the characteristics of GM( 1,1) and BP neural network model,the parallel grey neural network model( PGNN) is constructed and applied to forecast the settlement of a tunnel monitoring point of Nanjing metro. The results show that the prediction accuracy of PGNN is significantly higher than that of GM( 1,1) an...