Traversal grey model for water consumption prediction

Abstract

Grey model (GM) method is widely used to forecast water consumption due to the fact that it only needs little original data. To increase the precision of GM(1,1) and forecast urban water consumption accurately, we construct Traversal Grey Model (T-GM(1,1)) by minimum total residual sum of square. When we used the T-GM(1,1) to predict water requirement in some city, it showed that the mean absolute error, mean absolute percentage error and the maximum absolute percentage error are 0.77×10 6 m 3 /a, 3.56% and 12.58% respectively. Thus, the simulative precision of T-GM(1,1) is much better than that of GM(1,1). In summary, the T-GM(1,1) can predict the water consumption in the city and reduce the unstable factors that influence the forecast.