Evaluation of Seasonal, ANN, and Hybrid Models in Modeling Urban Water Consumption A Case Study of Rash City

Document Type : Research Paper

Authors

1 Assoc. Prof. of Agricultural Economy, Islamic Azad University, Marvdasht Branch, Marvdasht

2 Ass. Prof. of Agricultural Economy, College of Agriculture, University of Guilan

Abstract

Forecasting future water consumption in cities to plan for the required capacities in urban water supply systems (including water transmission networks and water treatment facilities) depends on the application of behavioral models of uban water consumption. Being located in the North-South corridor, Rasht City is assuming a new role to play in the national economy as a foreign trade center. It will, thus, be necessary to review its present urban infrastructure in order to draft the required infrastructural development plans for meeting the city’s future water demands. The three Seasonal Autoregressive Integrated Moving Average (SARIMA), Artificial Neural Network (ANN), and SARIMABP approaches were employed in present study to model and forecast Rasht urban water consumption using monthly time series for the period 2001‒2008 of urban water consumption in Rasht. The seasonal unit root test was applied to develop the relevant SARIMA model. Results showed that all the seasonal and non-seasonal unit roots are present in all the frequencies in the monthly time series for Rasht urban water consumption. Using a proper filter, the SAIMA patterns were estimated. In a second stage the SARIMA output was used to determine the ANN output and the hybrid SARIMABP structure was accordingly constructed. The values for Rasht urban water consumption predicted by the three models indicated the superiority of the SARIMABP hybrid model as evidenced by the forecast error index of 0.41% obtained for this model. The other two models of SARIMA and ANN were, however, found to yield acceptable results for urban water managers since the forecasting error recorded for them was below 1%.

Keywords

Full Text

شماره صفحات این مقاله از 89-84  به  98-93 تغییر کرده است.

References

1. Zhou, S.L., McMahon, T. A., Walton, A., and Lewis, J. (2002). “Forecasting operational demand for an urban water supply zone.” Journal of Hydrology, 259, 189-202.
2. Balling, R.C., and Gober, P. (2007). “Climate variability and residential water use in the city of Phoenix, Arizona.” Journal of Applied Meteorology and Climatology, 46, 1130-1137.
3. Arbués, F., Garcia-Valiñas, M.A., and Martínez-Espiñeira, R. (2003). “Estimation of residential water demand: A state-of-the art review.” J. Socioecon., 32, 81-102.
4. Statistical Center of Iran. (2011). Population and housing census, Tehran. (In Persian)
5. Franses, P.H., and Hobijn, B. (1997). “Critical value for unit root tests in seasonal time series.” Journal of Applied Statistics, 24, 25-47.
6. Taylor, A.M.R. (1997). “On the practical problems of computing seasonal unit root tests.” International Journal of Forecasting, 13, 307-318.
7. Ghiassi, M., David, K., Zimbra, K., and Saidane, H. (2008). “Urban water demand forecasting with a dynamic artificial neural network model.” Journal of Water Resource Planning and Management, 134(2), 138-146.
Journal of Water and Wastewater; Ab va Fazilab ( in persian )
Volume 27, Issue 4 - Serial Number 104
serial number: 104
September and October 2016
Pages 93-98

Files

Share

How to cite

Statistics