Skip to main content

Table 4 Models overall performance

From: Short-term water demand forecasting using hybrid supervised and unsupervised machine learning model

Model Model identifier MAPE (%) NRMSE (%)
1 h 8 h 24 h 7 days 1 h 8 h 24 h 7 days
SARIMA S-24 15.37 17.84 17.81 21.38 18.93 18.76 19.08 24.17
S-168 14.85 15.72 17.61 18.73 17.31 17.95 19.02 19.83
RT RT 11.48 12.93 13.72 16.75 12.43 13.19 17.81 21.04
Hybrid HYB-N2 08.62 09.58 11.41 13.89 10.83 12.83 14.47 18.62
HYB-N3 06.73 07.34 08.92 09.74 07.97 09.51 10.73 11.86
HYB-N4 04.97 06.03 06.57 06.93 05.54 06.86 08.73 09.83
HYB-N5 04.84 05.82 06.12 06.80 05.18 06.24 06.78 08.62